Control Interfaces for Intention Detection in Active Transfemoral Prosthetics: A Systematic Review

Control Interfaces for Intention Detection in Active Transfemoral Prosthetics: A Systematic Review | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Control Interfaces for Intention Detection in Active Transfemoral Prosthetics: A Systematic Review nur hidayah mohd yusof, Nur Hidayah Mohd Yusof, Nur Azah Hamzaid, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2814842/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Oct, 2025 Read the published version in Physical and Engineering Sciences in Medicine → Version 1 posted 6 You are reading this latest preprint version Abstract In this paper, a cutting-edge method for creating control interfaces for intention detection in active transfemoral prosthetic devices is presented. Given the current trend in the prosthetics industry, a review of the literature over the last two decades has found a number of control algorithms utilized for intention detection implementation. Scientific publications, books, and online resources were evaluated for published material on knee prosthesis. Based on the materials, three areas of scientific inquiry involving control interfaces for intention detection in active prosthetic legs have been identified. The studies were assessed using the Downs and Black checklist, and their control techniques as well as the performance assessment for these control interfaces are described. After screening, 211 studies were retrieved and examined, however only 39 publications were included and examined in this review. An active prosthetic leg's control strategy framework and goal output were examined in fifteen (15) papers. In two (2) papers, conventional control methods for transfemoral prosthetic legs were examined. Eight (8) further research looked at the potential implementation of intent detection in the transfemoral prosthetic leg, while fourteen (14) papers explored the active prosthetic leg's machine learning algorithm. As a result, our research showed that using a less complex sensory system to control an active transfemoral prosthetic limb is possible when paired with a creative approach and control algorithm that can translate the restricted sensor data into a larger set of relevant data. Therefore, an effective sensory system (practicality and quality of the sensory input) and intention detection algorithm were required for an active transfemoral prosthetic limb. It is considered a complex activity because of the connected biomechanics of human gait, which the brain governs unconsciously. No prosthetic device has been able to replace a human limb in the same way that the original one would operate due to the ongoing challenges in prosthetic technology. Even with something as simple as walking, there are several components to the difficulty that may be broken down into different study areas. No matter how little, this calls for input from many different domains of knowledge. Considerations for future development might be based on this current analysis of the intention-detecting control interface of active knee prosthetic devices and prosthetic technology. This evaluation of the literature includes not only a study of the viability of control interfaces that support intention detection for an active prosthetic limb, but it also suggests a framework for categorizing various works in the subject. active prosthetic device user-intention control algorithm sensory system intent recognition Figures Figure 1 Figure 2 1 Introduction Over the past 10 years, several electrical and sensory technologies have been included into prosthetic knee joint systems (Hargrove et al., 2013 ), mostly to create the best prosthesis that mimics the typical gait of a person. Three categories can be used to categorise contemporary transfemoral prosthetic devices: passive knee joints (Hafner and Askew, 2015 ; Celebi et al., 2013 ; Johansson et al., 2005 ); mechanical knee joints (Johansson et al., 2005 ; Borjian et al., 2008 ; Varol and Goldfarb, 2007a ; Hargrove et al., 2013 ); and active knee joints, either powered or (Segal et al., 2006 ; D. A Winter et al, and Ao et al, 1975; Song et al., 2008 ; Martinez-Villalpando and Herr, 2009 ). The market now offers only energetically passive transfemoral prosthetic devices that are commercially available. To offer enough ground clearance, the majority of passive device mechanisms restrict knee flexion during the stance phase and execute it during the swing phase. To prevent the knee from buckling and dropping, the user must fully extend their knee upon heel impact in order to lock the joint. Amputees must meet these two requirements in order to completely control a passive prosthetic knee joint. The knee flexion and extension angular velocities, however, may be changed during swinging by using a damper. Amputees can effortlessly manipulate their prosthetic legs thanks to this technology. Through the development of innovative solutions that help patients restore their regular stride, several studies have been conducted throughout the years to advance prosthetic technology. Using a microprocessor as a controller is one approach (Bellmann et al., 2010 ; Thiele et al., 2014 ). There are two categories of microprocessor-controlled prosthetic legs: mechanically passive and mechanically active (Hafner and Askew, 2015 ; Segal et al., 2006 ; Hafner et al., 2007 ; Sup et al., 2008a ). Both of these technologies may bring back a number of locomotor modes that analyse and interpret information from the prosthetic legs' sensory system using finite state machines, such as mechanically oriented sensors or physiologically oriented sensors (Hamzaid et al., 2020 ). There is currently a prosthesis available that can assess state changes in locomotor modes (Fite et al., 2007 ; Au et al., 2008 ). But to overcome the limitations of microprocessor-controlled prostheses, an intent detection approach was created to allow seamless, automatic, and natural transitions between locomotor modes (Young et al., 2014 ). This technique, as well as others, may be used to mechanically active or passive microprocessor-controlled prostheses. In light of this recent development, the goal of this study is to illustrate the various scientific research avenues that have been explored during the previous 10 years, with considerations on their experimental methods and findings through D&B analysis. Earlier research (Babuka and Verbruggen, 1996; Torrealba et al., 2008 ; Pieringer et al., 2017 ) reviewed the designs of various knee prosthesis, but this was performed by identifying the intent detection technique and classifying its control interface and approach. This study emphasises the key elements of the prosthetic leg control literature: sensors for intention detection, control strategies, and performance assessment methods. In each paragraph, the components or pieces needed to build a sophisticated knee prosthetic system are highlighted. Finally, a description of the evolution of prosthetic technology systems from the simplest to the most complex is given. 2 Methodology 2.1 Literature Search Strategy Within the previous 18 years, from May 2005 to June 2022, a literature search was carried out utilising the Google Scholar, Web of Science, IEEE Xplore, and PubMed databases. Based on the framework's investigation into intention detection, the year was selected. The phrases "sensory system," "controller," "gait intention recognition," and "microprocessor control" in combination with "active prosthetic transfemoral leg" were used to extract pertinent search terms, keywords, or publications. Studies that were reported in English and satisfied the qualifying requirements were kept for additional investigation. 2.2 Eligibility Criteria This review concentrated on transfemoral amputees' usage of prosthetic devices. The investigations were not just proof-of-concept research in the field, but also studies that reported prototypes of active prosthetic legs. Additionally, if the system was developed and tested on orthoses or upper limb prostheses or if the criteria had the potential to be applied to prosthetic devices generally, the "intention detection" criteria were also taken into account in this evaluation. Due to poor data presentation and result validation, conference papers and proceedings were omitted from this review. 2.3 Data Extraction from Selected Articles Four writers conducted this review (NH, NAH, LKW, and FJ). The titles and abstracts of the research were used to find those that could be eligible. We only kept and studied full-length papers that described the active control mechanism and user intention detection. Studies were included if the user intention and its sensory system were applied to prosthetics and orthotics as the primary component of the reported creation of the intervention of interest. Passive prosthetic devices were not included in the studies for this evaluation since the co-interventions might have influenced the findings or scope of the analysis. 2.4 Assessment of Study Quality The Downs and Black checklist were created to assess the methodological excellence of comparison research, both randomized and nonrandomized. The checklist's 27 elements include reporting, external validity, internal validity (bias and confounding), and power, which are all methodological components. One item was graded on a 3-point scale (yes = 2, partial = 1, no = 0), while the other 26 items were assessed as either yes (= 1) or no/unable to identify (= 0). Values range from 0 to 28, and higher scores indicate that the study's methodology was sounder. To classify research by quality, the following cut-points have been proposed: excellent (26–28), good (20–25), fair (15–19), and bad (14). The Downs and Black checklist have been shown to have adequate psychometric qualities, including internal consistency, test-retest reliability, inter-rater reliability, and criterion validity, in other publications. The top six quality evaluation techniques eligible for use in systematic reviews include the checklist. The Downs and Black (D&B) checklist for quality assessment's search approach was used to perform this systematic literature review (Downs and Black, 1998 ). The search strategy specifies the query, databases, and inclusion and exclusion standards for sorting the records that are obtained. The control interface tactics used by an active prosthetic limb and its sensory system are the major subject of this review. The D&B (AT, 2008 ) checklist methodological criteria were used to evaluate the review studies' quality. Some of the criteria were disregarded since they applied solely to randomized controlled trials and were not relevant to the current development of prostheses. Examples of items that were not applicable in this review were those that dealt with external validity (items 12 and 13), internal validity (items 14, 15, and 16), and internal validity - confounding selection bias (items 21, 22, 23, and 24). In order to compare the methodological quality of the included studies, the D&B score was normalized based on the grading scale of the inclusion and exclusion criteria of the study. 2.5 Classification Table To obtain an overview of all available prosthetic knee devices, the required criteria or components are summarized in Table 1 . The main characteristics of the different systems were structured according to the following criteria: Table 1 Summary of control frameworks strategies for an active prosthetic leg Control strategy Control modes Actuation principles Sensory systems Classical control (El-Sayed et al., 2014 ; Furuya et al., 2013 ) Classical Electro-mechanical Mechanical Machine learning technique (Bellmann et al., 2010 ; Alzaydi et al., 2011 ; Kadhim et al., 2020 ) Intelligent Pneumatic & powered Mechanical Pattern recognition (Young et al., 2014 ; Huang et al., 2011 ; Brantley et al., 2017 ; Shaikh and Malhotra, 2020 ) Intelligent Hydraulic & powered Mechanical & biological Volitional control (Dawley et al., 2013 ; Bai et al., 2016 –; Ha et al., 2011 ) Intelligent Hydraulic & powered Mechanical & biological Myoelectric control (Young et al., 2014 ; Aeyels et al., 1995 ; Delis et al., 2009 ; Huang et al., 2010 ; Wu et al., 2011 ) Intelligent Hydraulic & powered Biological Echo control (Varol and Goldfarb, 2007b ) Intelligent Hydraulic & powered Biological Prosthetic knee joint: This acts as an identifier for the prosthetic knee joint system. Sensory system and signal: This indicates the types of sensors used in the transfemoral prosthetic leg and the signal acquired. Actuator characteristics: This indicates the characteristics of the actuator used in the prosthetic leg concerning the motor, transmission, power rating of the actuator, and moment rating at the knee joint. Control strategies: This indicates the control strategies used in the prosthetic leg to connect both sensory and actuator systems so that seamless transition of locomotion modes can be achieved. Stage: This indicates whether the prototype is in the lab experiment stage, simulation, platform test design, or is already commercialized in the market. Performance assessment tools: Indicates the tools or equipment used in the studies to assess or analyze the performance of the active prosthetic leg. Ambulation tasks: This indicates whether the prosthetic knee device can perform all ambulation functions, such as level ground walking, stairs and ramp negotiation, and sit to stand. User intention detection, which indicates the intention detection ability of an active transfemoral prosthetic leg. Number of subjects: This is the number of subjects that participated in the study. 3 Results And Discussion 3.1 Characteristics of Included Studies Initial selections for inclusion included 17495 articles (Fig. 1 ). After eliminating any duplicate or related papers following title and abstract screening, only 211 studies were retrieved and examined. After a thorough evaluation, 179 articles were eliminated since the inclusion requirements weren't met. Last but not least, this review included, evaluated, and examined 39 papers (Fig. 1 ). An active prosthetic leg's control strategy framework and goal output were examined in fifteen (15) trials. In two (2) trials, conventional control methods for transfemoral prosthetic legs were examined. Eight further research looked at the potential implementation of intent detection in the transfemoral prosthetic leg, while fourteen (14) papers explored the active prosthetic leg's machine learning algorithm. 3.2 Methodology Quality The control interface mechanisms for intention detection that may be used in an active prosthetic limb are realistically scientifically supported in this work. The quality of the studies varied from acceptable to good, with scores between 40 and 100% (Table 4 ), with a mean score of 83.3%, which is regarded as excellent. This is according to the accepted D&B checklist (AT, 2008 ). All studies accurately described the research's purpose and underlying premise (criteria 1) as well as the reliability of its outcome measures (criteria 20). Overall, the results of these investigations came from trustworthy lab tests. The paucity of application of the probability value results indicates that few of them provided clinically pertinent data. Most D&B criterion items demonstrate how few finds and advances are generally practical. The papers under consideration provided evidence with acceptable methodological quality that other researchers may use as a guide or a point of reference. The majority of these studies, with the exception of Dawley et al. (Dawley et al., 2013 ) and Johansson et al., mentioned the result measured in their methods section (criterion 2) (Johansson et al., 2005 ). All authors with the exception of Johansson et al. clearly explained the study's intervention (criterion 4) (Johansson et al., 2005 ). Additionally, every study—aside from Dawley et al. (Dawley et al., 2013 –clearly explained the study's primary results) (criteria 6). With the exception of Varol et al, all studies adjusted for confounding in the analyses from which the main conclusions were mostly not presented (criterion 25) (Sup et al., 2007). 3.3 Sensors in Transfemoral Prosthesis for Intention Detection Throughout the years, several questions have been raised to develop a good intent recognition system that can provide the best user intention recognition with fast and reliable predicted performance. Previous studies have addressed these questions in different ways. The mechanism of the sensory system can be addressed using a wide range of approaches and varying degrees of invasiveness. Generally, two types of sensory systems have been adopted for user intention detection: prosthetic device-oriented (Sup et al., 2008b ; Sup et al., 2007a ; Varol et al., 2010 ; Young et al., 2014 ; Yusof et al., 2018) and biological-oriented (Furuya et al., 2013 ; Alzaydi et al., 2011 ; Bai et al., 2016 ). Prosthetic devices with user-oriented sensors primarily use mechanical sensors to detect the biomechanics of gait, such as the measurement of forces, torques, joint angles, and vertical orientation. For example, Sup and team (Sup et al., 2008b ) used a load cell placed between the prosthesis and the user to derive joint torque and motion. The sensors measured the interaction forces and moments between the prosthesis and the user for control and intent recognition purposes. In addition, Sup et al. (Sup et al., 2007) used mechanical sensors mounted on a powered prosthesis as a control algorithm to predict user intention. On the other hand, Jasni and colleagues (Jasni et al., 2016 ; Jasni et al., 2019 ) used piezoelectric sensors as in-socket sensory systems for intention detectors in active transfemoral prosthetic legs. The studies reported the efficacy of piezoelectric sensors to be used as a potential input signal for the control system of an active prosthetic leg using force profiles of muscle pressure on the socket and the corresponding ground reaction force while performing movements. In addition, El-Sayed and his team (El-Sayed et al., 2014 ; Mohd Yusof et al., 2018 ) utilized piezoelectric transducers as a feedback signal source in the simulated control of an active prosthetic leg. Researchers have used movement characterizations based on sensor sensitivity against the stump, and thus provided an accurate voltage signal to the control feedback system. Although these studies have demonstrated the reliability of piezoelectric sensors for use as an in-socket sensory system for user movement and intention detection, they have yet to verify that such a signal can be used as an input to a feedback control system. To overcome the shortcomings of prosthetic device-oriented sensors, user-biological-oriented sensors have also been adopted. Surface EMG sensors are one of the major neural-control sources for prosthetic leg development that provide information for intention detection purposes from the muscle contractions in the amputee’s residual limb, and then give commands and modulate the output of the prosthesis. 3.4 Control Strategies for Actuation System of an Active Prosthetic Leg The purpose of having a control system is to execute the intended movement of the prosthetic knee and ensure a seamless transition mode during locomotion activity. The control system architecture of an active transfemoral prosthetic leg can be described as a three-level hierarchical controller scheme (Khademi et al., 2019 ) as illustrated in Fig. 2 . The high-level controller acts as an intent recognizer, and an intelligent control scheme was used at this level, consisting of three sub-parts: locomotion mode detector, slope calculator, and cadence calculator. The locomotion mode detector differentiates the ambulation modes of the user and sends the information to the middle-level controller, whereas booth calculators estimate the required amount of slope and cadence during walking. The middle-level controller controls the ambulation function mode (e.g., walking, slope, sit-to-stand, and climbing stairs) of the system, whereby it generates the required torque reference for the joints by modulating the variable knee impedance depending on the phase of the ambulation mode using a finite state machine. This intelligent control system monitors the movements and inputs of amputees and generates commands that the prosthesis actuator should perform. At the lowest level of the control hierarchy, a classical control (CC) scheme was used for transmission dynamics or feedback purposes. The intelligent control system is normally used in high-level controllers to ensure a reliable interaction with amputee users as well as to provide continuous transition between locomotion patterns. 3.5 Low Level Controller (Classical) Various CC schemes can be used as a controller system architecture for active prosthetic legs, depending on the type of prosthetic leg to be developed. The common method is a proportional integral derivative closed- and open-loop control. Proportional integral derivative control is a way of driving a system towards a target position or level, using a control loop feedback mechanism to control process variables. Normally, the CC strategy (El-Sayed et al., 2014 ) is used to model the dynamic behavior of the prosthetic system itself before developing the actual one, and this strategy is used in a low-level controller. Researchers used this control method to construct a mathematical model of the system before implementing the actual model. Thus, the safety of the prosthetic leg is ensured. It is recommended to be used as a first-stage process to determine the kinematics and kinetic behavior of lower-limb prostheses. Classical control CC can be performed using MATLAB or any other simulation software, which is based on dynamic-mathematical modeling and is considered extremely complex but less robust. Thus, this type of control system is difficult to implement in real-time (Zlatnik, 1998 ). El-Sayed et al. (El-Sayed et al., 2014 ) used this strategy to analyze the effectiveness of the actuation system of prosthetic legs, and the results presented a realistic simulation in terms of knee parameters. However, its real-time performance requires further verification. 3.6 Middle Level Controller In the middle-level controller, finite-state impedance control is used as a soft control scheme to emulate the impedance behavior of the healthy biomechanical gait. This control strategy was based on pre-stated rules that combine the information of certain knee parameters, thus allowing the control action of the prosthetic actuator system. This impedance strategy is emulated by modulating the impedance of the prosthetic knee joint according to the phase of the gait. Each gait phase had parameters for stiffness, damping, and spring equilibrium angle for the knee joint. In addition, the purpose of this finite state control strategy is to enable interaction between the user and prosthesis. That is, in any given state, the behavior is passive and would come to rest at local equilibrium, thus providing reliable and predictable behavior for the user. Zlatnik (Zlatnik, 1998 ) used this strategy as the knowledge base for a system that stored the predefined rules and parameters required to facilitate the knee joint in all types of ambulatory activities. The results of this study prove that this control method can predict the performance of a prosthesis and produce the required moment. Sup et al. (Sup et al., 2008b ; Sup et al., 2007a ) used this approach to manipulate the required torque at each joint during a single stride, represented as a piecewise series of impedance functions. For example, the stance flexion behavior of the knee can be approximated as a linear spring, whereas the pre-swing knee torque is described as a nonlinear spring. Thus, one can coordinate the prosthesis motion and control-level walking gait. Varol et al. (Varol et al., 2010 ), also used a finite-state impedance strategy to execute the joint impedance according to the gait phase. In each phase, the knee and ankle torque was described by a passive spring and damper with a fixed equilibrium point and by shifting the joint impedances between the gait phases. For instance, shifting from a late state during walking to swing flexion occurs with the detection of toe-off. This structure indicates that the prosthesis is guaranteed to be passive within each gait phase and delivers power to the user only by shifting the linear stiffness and equilibrium point of the virtual spring between the phases. Therefore, prosthetic users can control prostheses directly and naturally. The results from this study demonstrated the robustness of the control system approach, which can be implemented in an active prosthetic leg for the user to perform various ambulatory movements more naturally. 3.7 High Level Controller (Intent Recognition) A suitable input signal and fast machine-learning algorithm are required to design a good and reliable intent recognition system. Researchers have used various techniques to address this issue (Table 1 ). Determining the proper state transition within the locomotion modes (e.g., between the stance and swing phases of walking) is easier than defining the state transition between the locomotion modes (e.g., from the stance phase of walking to the swing phase of stair ascent), which is the reason why a high-level controller was developed. The purpose of having a high-level controller is to predict the intention of the prosthetic user, and this intent recognition is achieved through a pattern recognizer that compares the state of the prosthesis to probabilistic models of activity. Misclassification of intent recognition can cause imbalances among users. Several intent recognition strategies have been developed by researchers to be used as intent recognizer systems and are presented below. 3.8 Locomotion Mode Recognition Strategy A more robust and accurate intent recognition system can be developed by classifying specific points during a single gait cycle. Previous studies by Young et al. (Young et al., 2014 ) and Varol et al. (Varol et al., 2010 ) introduced data classifier windows based on each gait phase. A model of the group data was trained based on the sets of selected input sensors. Thus, the data were extracted to a certain frame length and converted into appropriate sets of features to be extracted from each window. After the input was selected, a set of training data was used to develop the model for each locomotion mode: walking, sitting, or standing. The model was established in real-time to determine the exact activity mode at a given instant. This strategy was proven successful but required a low-pass filter to improve the current mode determination in real-time. In contrast, Khademi et al. (Khademi et al., 2019 ) used a gait mode recognition approach to provide maximum performance with minimum complexity for transfemoral amputees to control their prostheses. The purpose of introducing the new framework was to design a comprehensive control system that utilized multi-objective optimization. This was to find the optimal feature subsets that can produce an intention recognition system that is both precise and accurate. The results reported a 97% classification accuracy in detecting the locomotion mode with six subjects. Therefore, this study indicated the reliability and compactness of advanced optimization methods for locomotion mode detection implemented in active prosthetic legs. Yusof et al. (Yusof et al., 2018) proposed user gait mode recognition based on seven phases of classification from the in-socket sensory system that allows one to mimic a normal healthy walking gait. The purpose of the in-socket sensory system was to derive the user’s intention to actively actuate the knee and thus reduce the user’s metabolic energy consumption. A control framework approach using an adaptive neuro-fuzzy inference system (ANFIS) algorithm was proposed to decode the arrays from the in-socket sensory system and characterize it into the seven phases of gait states from heel strike to toe-off. Subsequently, the particular gait phase recognition was mapped to the knee joint actuator torque and cadence control output. The proposed control framework was validated using in-socket sensor signals from 30 gait cycles of a transfemoral amputee. The results showed 80–90% accuracy in terms of gait kinematics and kinetic parameters. The ANFIS system has been proven to detect all seven phases of gait based on the amputee’s in-socket sensory system. However, although this study presented the reliability of the ANFIS control strategy as a control algorithm framework for user movement and intention detection, the authors have yet to verify its implementation in real prosthetic devices. 3.9 Myoelectric Control Strategy Researchers have investigated and implemented surface electromyography (EMG) for intention detection to control the knee joint of a transfemoral prosthetic leg. Aeyels et al. (Aeyels et al., 1995 ) developed a computer-based control system using EMG surface sensors and an electrically modulated brake for gait mode identification. The EMG sensors act as switches or moderators to provide commands to the actuator system for the required gait mode. The major advantage of using EMG-based control is the ability of the amputee to move the prosthesis according to their desire, instead of limiting the motion to a certain point of locomotion function (Varol and Goldfarb, 2007a ; Young et al., 2014 ; Huang et al., 2011 ; Furuya et al., 2013 ; Alzaydi et al., 2011 ; Bai et al., 2016 ; Delis et al., 2009 ; Huang et al., 2010 ). Kevin et al. (Ha et al., 2011 ) used a volitional control strategy during non-weight-bearing activities that utilized an impedance framework as its algorithm scheme. The joint was programmed with a given stiffness and damping value, which reflected the nominal impedance properties of an intact joint. Surface EMG placed on the user’s hamstring and quadriceps muscles of the residual limb acted as the set-point angle of the joint impedance. Rather than using EMG sensors only to command the flexion and extension of the knee, this study used a combination of quadratic discriminant analysis and principal analysis to align the user’s intent to flex or extend the knee joint. However, this approach is limited to non-weight-bearing activities; thus, it is not suitable for other locomotion settings. Wu et al. (Wu et al., 2011 ) proposed the use of an EMG-based control approach for amputee users to directly control the active prosthetic leg using their muscles by activating the neural signal between the EMG sensors and the muscle itself. This mechanism of ‘active-reactive’ control strategies mimics the actuation mechanism of a human biological joint. In addition, this study investigated the compatibility between EMG sensors and the amputee user’s skin while performing various strategies. Surface EMG sensors proved to be a non-intrusive interface for acquiring a user’s muscle-activating signals. The findings from the study indicated that the controller could predict the knee and ankle joint positions with the user’s neural command, which was similar to the standard gait of healthy subjects. Brantley et al. (Brantley et al., 2017 ) on the other hand, utilized EMG sensors to extract the signal from the amputated limb using a nonlinear extension of the Kalman filter to infer the users’ intended gait pattern and thus predict the kinematic values of the knee and ankle joints. Thus, neural machine interfaces can provide a more natural and continuous transition of gait during locomotion activities. This study specifically used the anterior and posterior muscles as indicators to predict the ankle and knee positions, which resulted in high accuracy. Moreover, this study provides insight into the improvement of closed-loop control of lower-limb prostheses when compared with other intent-recognition strategies. Despite all the interventions has been made in the prosthetic technology field, there is yet to be a commercialized prosthesis available in the market that utilizes EMG sensors. According to Aeyels et al. (Aeyels et al., 1995 ), there are several difficulties in implementing these strategies into commercialized ones, including identifying and capturing the EMG sensor signals, processing them, and characterizing them according to the prosthetic device orientation. The effects of muscle fatigue could also cause additional problems in the control system of the prostheses. 3.10 Echo Control Strategy The echo control strategy uses a sound limb kinetic profile that is modified to control the prosthesis side. Sensors were instrumented on the intact limb of unilateral amputees to capture its knee angle profile and adapted for use on the powered prosthesis side one-half cycle later. This was referred to as “modified trajectory echo control” in the development of an electro-hydraulically powered knee prosthesis (Varol and Goldfarb, 2007b ; Flowers, 1971 ). The prosthetic limb mimicked the kinematics of the sound limb through minimal system training. This method required users to train and control their gait at an even number of steps. Ossur (OSSUR, USA) developed a commercialized self-contained powered knee prosthesis, which featured a concept similar to that of echo control, in which instrumented sensors were embedded on the sound limb. One drawback of this technique is that the sound-side leg must be instrumented with sensors. This requires users to don and doff additional instrumentation, which can be challenging for them to do. To don and doff the prosthetic leg itself requires considerable work and energy; thus, this technique is not practically convenient for amputee users. Although this technique is restricted to unilateral amputee users, it also requires the user to keep track of an ‘odd’ number of steps, whereby an echoed step is undesirable. This forces the amputee to react to the limb, instead of interacting with it. 3.11 Volitional Control Strategy This control framework uses both mechanical and biological sensors to operate and is referred to as the biological-mechanical fusion system. Varol et al. (Varol et al., 2010 ) utilized a Gaussian mixture model as its control framework to characterize the probability of the user and the prosthesis being engaged in a given activity mode. EMG sensors were used as the signal transmitter between the user and control system to generate complex classifier models that were simple for various locomotion modes (sitting, walking, standing, slopes, and climbing stairs). In addition, the control algorithm also generates the equilibrium point of angular velocity and joint impedance, which consists of a combination of joint stiffness and damping. Therefore, the knee can move accordingly with the correct joint output stiffness and damping to the desired position, thus providing a smooth transition of the gait between the user, prosthesis, and environment. These Gaussian mixture model classifiers behave similarly to linear discriminant analysis and quadratic discriminant analysis classifiers. Similarly, Huang et al. (Huang et al., 2011 ) developed an algorithm based on neuromuscular-mechanical fusion to continuously recognize a variety of locomotion modes by prosthesis users. In the study, they used a support vector machine to decode the interfaces between the EMG sensors and user, thus providing the necessary data to the control system. The support vector machine technique has also been reported as a reliable classification method that provides better classification performance than any other artificial neural network for intended purposes. The findings from both of these studies can be used as guidance to further develop a volitional control framework for intention detection purposes of transfemoral prostheses. 3.12 Performance Assessment Another research line regarding the development of an active prosthetic leg is the performance validation of the prostheses. This could be related to their mobility capability in producing smooth transitions and the reliability of the prostheses in terms of performance and ambulation tasks, and an assessment of every prosthesis developed is required. Various assessment methods can be used to validate the performance, one of which is the use of motion analysis lab facilities, such as Vicon Nexus (Vicon Motion Systems Ltd., UK). A gait laboratory system provides detailed measurements of several human-body motion variables. This analysis allows researchers to obtain technical conclusions regarding the assessment of the prosthetic leg itself. Moreover, a gait lab can capture various joint movements. The reconstruction of the gait cycle is also possible. For instance, in prosthetic leg assessments, all angular parameters of the joint must be measured along with the ground reaction forces. The results obtained might be compared to the normal human leg database to determine how far the prosthetic leg performance is from an actual one with very high accuracy. Several studies have reported the use of gait analysis to study the performance of prostheses. For example, Jasni et al. (Jasni et al., 2016 ; Jasni et al., 2019 ) used gait lab analysis to study the efficacy of the sensory system in a prosthetic leg socket. The results of this study verified the reliability of the piezoelectric sensors as an alternative control input using the force profile of the muscles and the corresponding ground reaction force while performing the movements. In addition, El-Sayed et al. (El-Sayed et al., 2014 ) employed a gait lab to evaluate the performance of the transducer sensor feedback signal in controlling the movement of the prosthesis. In other words, both of these studies verified the feasibility of the sensors to be used as the input source signal for the control system of potential active prostheses by providing accurate signals of the voltage arrays. The studies presented and discussed the gait analysis results obtained from a transfemoral amputee who used their in-socket sensory system. Another way of assessing prosthetic leg performance is through comparison with other similar technologies. In this sense, Hafner et al. (Hafner et al., 2007 ) and Johansson et al. (Johansson et al., 2005 ) conducted a comparative study to assess gait patterns and variable damping between mechanical and microprocessor knee devices. In these studies, metabolic data were collected from eight unilateral amputees while walking with the selected knee device to analyze the kinetics and kinematics of the prosthetic devices. A comparison of the metabolic rate energy of the users was also performed in these studies. This means that the metabolic consumption when using different prosthetic knee joints was measured and the efficiency of the design was obtained. Simulation studies have also been used as one method to assess the behavior of prosthetic legs. Yusof et al. (Yusof et al., 2018) used this method to validate a control algorithm for an active prosthetic leg. The purpose of using physical simulation is to estimate the dynamic behavior of the system and simulate the movement of the prosthetic leg. To begin the development of an actuated knee joint, simulation is recommended as the first-stage process to analyze whether the design or control framework can produce the desired ambulatory parameters. El-Sayed et al. (El-Sayed et al., 2014 ) used this method to model and control the linear actuated transfemoral knee joint in basic daily movements. The simulation results presented a realistic simulation of the actuated mechanism in terms of the knee parameters. Other studies have gathered subjects’ opinions for setting conclusions rather than using gait analysis alone to validate the performance of the prosthetic device. The qualitative results may provide better depth with regard to the users of the prosthesis itself. On the other hand, studies reported by Andrysek et al. (Andrysek et al., 2005 ) evaluated the conceptual basis of the stance-phase controlled pediatric prosthetic knee joint design through a clinically tested prototype and used a questionnaire to analyze the efficacy of the knee system. Kirker et al. (Kirker et al., 1996 ) also used this method to evaluate the performance of different types of intelligent prosthetic knee devices. All prosthetic knee devices were assessed using a questionnaire given to six patients. Table 2 summarizes the performance assessment tools used for the development of an active prosthetic leg. Table 2 Summary of performance assessments tool for development of an active prosthetic leg Performance assessments validation Assessment tools Advantages Limitations Gait analysis lab (Jasni et al., 2016 ; Jasni et al., 2019 ) Vicon Nexus All the kinetics and kinematics parameter can be tested Limited to indoor activity testing Comparative study (Hafner et al., 2007 ; Johansson et al., 2005 ) Metabolic data with various variables Easy to trace the problems in the prosthetic device and troubleshoot it Limited to commercialized prosthetic devices in the market Simulation study (El-Sayed et al., 2014 ; Mohd Yusof et al., 2018 ; Bhakta et al., 2020 ) MATLAB software, OpenSim simulations Can simulate the movement of the device before fabrication stage, durable and safe Limited to software analysis Survey analysis (Andrysek et al., 2005 ; Kirker et al., 1996 ; Saglam et al., 2017 ) Questionnaire Easy to handle Results might be less scientific and valid, time consuming 3.13 Classification Table Various control strategies can be implemented in prosthetic devices to achieve intention detection ability. Table 3 provides a summary of previous studies conducted by other researchers regarding the criteria that must be considered to produce an active prosthetic leg. Table 3. Summary of previous studies regarding the criteria needed for an active prosthesis development Author Prosthetic knee joint Sensory system & signal Actuator characteristics (Types of motor Power rating Moment) Control strategies Stage Performance assessments tools Ambulation tasks User-intention detection No. of subjects Ammar et al. (Alzaydi et al., 2011) Active prosthetic knee fuzzy logic Wireless (EMG and accelerometers) DC brushed motor 7648 rpm Artificial neural network Lab experiment/ simulation/ Platform test design Platform test design Walking only Yes None Bellman et al. ( Bellmann et al., 2010) Microprocessor knee (C-leg, Hybrid knee, Rheo knee, Adaptive 2) Mechanical sensors Hydraulic actuator Finite state machine variable damping Commercial in the market Gait laboratory (Optoelectronic camera system combined with 2 force plates) Yes 10 Varol et al. (Varol et al., 2010) Powered knee prosthesis Load cells Electric motor, lithium battery 118 W Locomotion mode recognition + finite state-based impedance Lab experiment Gait lab analysis Walking, sitting, and standing Yes 1 Young et al. (Young et al., 2014) Powered prosthesis Accelerometers + gyroscopes + load cells Robotic leg built by Vanderbilt University Finite state machine control Lab experiment Gait lab analysis Walking, sit to stand, ramp and slope Yes 6 Varol et al. (Varol et al., 2009) Powered prosthesis Load cells Electric motor Gait mode intent recognition Lab experiment Gait lab analysis + simulation analysis for control system purposes Walking at different speeds, standing Yes 1 Johansson et al. (Johansson et al., 2005) C-leg, Rheo knee, Mauch knee EMG + accelerometers Hydraulic motor Finite state control Lab experiment+ already commercialize in the market Gait lab experiment Ground level of walking Yes 8 Dawley et al. (Dawley et al., 2013) Powered knee with impedance control EMG+ uniaxial load cell Dual regenerative servo amplifiers Impedance-based control Lab experiment Gait lab analysis Walk in multi-terrain Yes 6 Huang et al. (Huang et al., 2011) Hydraulic passive knee EMG + load cells Hydraulic servo Pattern classification Lab experiment Gait lab analysis Walking, stair ascent and descent, obstacles, ramp Yes 6 Young et al. (Young et al., 2013) Powered knee prosthesis 16 mechanical sensors (load cells, potentiometers, encoders) Double acting pneumatic actuator, four-way servo valves, Impedance based control using time history information Lab experiment Gait lab analysis Ground level walking, ramp walking 10-degree slope, ascend descend stairs Yes 6 Sup et al. (Sup et al., 2007b) Powered knee prosthesis Load cells, uniaxial load cells, joint motion sensors Double acting pneumatic actuator, four-way servo valves, lithium battery Impedance based control approach Lab experiment Gait lab analysis Walking at a different speed Yes 1 Jasni et al. (Jasni et al., 2016) Hydraulic knee prosthesis Piezoelectric sensors Hydraulic actuator (OTTOBOCK) Impedance based control Lab experiment Gait lab analysis Walking different speed Yes 1 El-Sayed et al. (El-Sayed et al., 2014) Powered knee - - Variable impedance control approach Simulation software Simulation software Walking at a different speed Yes None Wu et al. (Wu et al., 2011) Active above-knee prostheses EMG + joint angle (uniaxial load cell + rotary potentiometer) DC motor+ ball screw assembly 150 W Echo + impedance Lab experiments/ lab testing Set of free swing experiments + set of level walking experiments Yes Ha et al. (Ha et al., 2011) Powered knee EMG, uniaxial load cells (ELPF-500L), potentiometer. (ALPS RDC503013), moment sensor Maxon EC30 Powermax brushless motor, 200 W, lithium polymer battery, 29.6 V, nominal rating 4000 mA.h capacity Impedance-based weight-bearing control Lab simulation MATLAB real-time workshop Sit to stand only Yes 3 Khademi et al. (Khademi et al., 2019) OTTOBOCK prosthesis Sensors from the Vicon system camera OTTOBOCK actuator system Gait mode recognition control Lab experiment + simulation analysis Gait lab analysis using 16-camera Vicon Nexus Ground walking, standing, and walking at various speeds. Yes 6 Varol and Goldfarb (Varol and Goldfarb, 2007a) Powered knee Load cells Servo-valve Locomotion mode recognition + finite state-based impedance Lab experiment Gait lab analysis Walking, sitting, and standing Yes 1 Brantley et al. (Brantley et al., 2017) Powered knee Not stated Not stated Finite state control approach Lab experiment Gait lab analysis Walk on a multi-terrain gait course (stairs, ramp, and ground-level) Yes 6 Delis et al. (Delis et al., 2009) Micro-controlled prosthesis EMG, accelerometers, Kalman filter. Hydraulic Pattern recognition Lab experiment Gait lab analysis & MATLAB simulation Walking Yes 4 able-bodied individuals Grimmer et al. (Grimmer et al., 2014) Active prosthesis Force sensors, infrared cameras Series elastic actuator (SEA) with spring Finite state impedance control Lab experiment Gait lab analysis Walking and running Yes 28 Yusof et al. (Yusof et al., 2018) Powered prosthesis Piezoelectric sensors Hydraulic knee Finite state impedance control (ANFIS) Lab experiment Gait lab analysis Walking at different speeds Yes 1 Bai et al. (Bai et al., 2016–) Powered knee prosthesis EEG and switch Single axis knee and manual locking system Volitional control of a knee-locker switch in real time Lab experiment Gait lab analysis Short distance of ground level walking Yes 1 Thatte and Geyer (Thatte and Geyer, 2016) Active knee SEA unit IMU and load cells Hybrid neuromuscular model Impedance Simulation experiments Gait lab analysis Walking No 1 Hargrove et al . (Hargrove et al., 2013) Powered prosthesis EMG Powered knee by Vanderbilt- University Volitional impedance control Lab experiment Gait lab analysis Sitting only Yes 12 El-Sayed et al. (El-Sayed et al., 2014) Mechanical prosthesis Piezoelectric Mechanical knee - Lab experiment Gait lab analysis Walking Yes 1 Pfeifer et al. (Pfeifer et al., 2015) Tethered powered knee prosthesis Load cells Visco elastic actuator (SVA), Maxon EC30 4-pole Finite state control strategy Lab experiment Pilot experiment Walking at different speeds No 1 Hargrove et al. (Hargrove et al., 2013) Powered prosthesis EMG Maxon EC30 Pattern recognition strategy Lab experiment Virtual environment experiment Sitting Yes 6 Jasni et al. (Jasni et al., 2019) Hydraulic knee prosthesis Piezoelectric sensors Hydraulic actuator (OTTOBOCK) Impedance based control Lab experiment Gait lab analysis Walking different speed Yes 1 Wu et al. (Wu et al., 2021) Powered knee prosthesis Potentiometer+ EMG a slider-crank mechanism with electric DC motor Reinforcement learning algorithm. Finite state machine + impedance control + intact knee motion tracking Lab experiment Gait lab analysis & OpenSim simulations Ground-level-walking Yes 2 able- bodies Shaikh & Malhotra (Shaikh & Malhotra, 2020) Micro-controlled prosthesis (Microcontroller LPC2148) Sensory feedback system IMU + 3-axis gyroscope linear actuator Motion Fusion Algorithm with real-time feedback. chip (PSoC 4), 042-BLE, with Bluetooth Lab experiment Gait lab analysis Different speed of walking Yes 6 able-bodied individuals 14 TF Wen et al. (Wen et al., 2020) Personalized robotic knee prosthesis load cell+ angle sensor DC motor Approximate Dynamic Programming (ADP) +Finite-state machine impedance control Lab experiment Gait lab analysis & OpenSim simulations Ground-level-walking No 1 able-bodied individual 1 TF Kadhim et al. (Kadhim et al., 2020) Powered knee prosthesis axis force sensors+ pressure sensors platform electric DC motor Adaptive Neuro-Fuzzy Inference System (ANFIS) Lab experiment Gait lab analysis Ground-level-walking Yes 1 Liu et al. (Liu et al., 2017) powered prosthetic knee EMG (MA300-XVI)+ load cell (Mini58, ATI, Apex, NC, USA) PAMs linear actuator pattern recognition (PR) + LMR (locomotion mode recognition) + finite-state impedance controller Lab experiment Gait lab analysis Level walking and ascending/descending slopes. Yes 2 able-bodied individuals 2TF Bhakta et al. (Bhakta et al., 2020) Powered knee and ankle prosthesis load cell+ embedded sensor encoders 2 brushless DC motors + gear transmission+ harmonic drive+ Li-Po battery impedance control + Finite-state machine impedance control+ Pattern recognition strategy Lab experiment Gait lab analysis & MATLAB data analysis Level walking and ascending/descending slopes. Yes 6 Table 4. Methodology quality of included studies according to downs and black checklist Study Criteria 1 2 3 4 5 6 7 8 9 10 11 12 13 16 17 18 19 20 25 27 Quality score Percentage (%) Bellmann et al. (Bellmann et al., 2010) 1 1 0 1 0 1 0 NA 1 NA NA NA NA 1 NA 1 1 1 0 1 10/14 71.4 Ammar et al. (Alzaydi et al., 2011) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Delis et al. (Delis et al., 2009) 1 1 1 1 1 1 1 1 1 1 1 1 NA NA NA 1 1 1 0 1 16/17 94.1 Grimmer et al. (Grimmer et al., 2014) 1 1 1 1 1 1 1 NA NA NA 1 1 NA NA NA 1 NA 1 0 1 12/13 92.3 Pfeifer et al. (Pfeifer et al., 2015) 1 1 1 1 1 1 1 1 1 1 1 NA NA NA NA 1 1 1 0 1 15/16 93.8 Hargrove et al. (Hargrove et al., 2013) 1 1 1 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 14/15 93.3 Yusof et al. (Yusof et al., 2018) 1 1 1 1 NA 1 1 1 1 1 1 1 1 NA NA 1 1 1 0 1 16/17 94.1 Varol et al. (Varol et al., 2010) 1 1 1 1 NA 1 1 1 1 1 1 1 1 1 NA 1 0 1 0 1 17/19 89.5 Young et al. (Young et al., 2013) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 NA 1 1 1 0 1 18/22 81.8 Young & Hargrove (Young & Hargrove, 2014) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 NA 1 1 1 0 1 18/21 85.6 Thatte & Geyer (Thatte & Geyer, 2016) 1 1 0 1 NA 1 1 1 1 NA NA NA NA NA NA 1 0 1 0 NA 9/12 75.0 Dawley et al. (Dawley et al., 2013) 1 0 1 1 NA 0 NA 0 NA NA NA NA NA NA NA NA 1 1 NA NA 5/8 62.5 Young et al. (Young et al., 2014) 1 1 1 1 0 1 1 0 1 0 NA NA NA NA NA 1 1 1 NA 0 10/14 71.4 Varol et al. (Varol et al., 2009) 1 1 0 1 1 1 1 1 1 1 NA NA NA 1 NA 1 1 1 1 1 15/16 93.8 Fite et al. (Fite et al., 2007) 1 1 0 1 0 0 0 1 1 0 NA NA NA 1 NA 0 0 1 0 1 8/17 47.1 Brantley et al. (Brantley et al., 2017) 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 1 1 NA NA 17/19 89.5 Jasni et al. (Jasni et al., 2016) 1 1 1 1 1 1 1 1 1 1 1 1 1 0 NA 1 1 1 0 1 17/19 89.5 Kevin et al. (Kevin et al., 2011) 1 1 1 1 1 1 1 1 1 1 NA NA NA 1 NA 1 1 1 NA 1 15/15 100.0 Delis et al. (Delis et al., 2009) 1 1 1 1 1 1 1 1 1 1 NA NA NA 0 NA 1 1 1 0 1 14/16 87.5 Varol & Goldfarb (Varol & Goldfarb, 2007a) 1 1 1 1 NA 1 NA NA NA NA NA NA NA 1 NA 1 1 1 0 1 10/11 90.9 Bai et al. 2015 (Bai et al., 2016) 1 1 1 1 1 1 1 1 1 1 NA NA NA 0 NA 1 1 1 0 1 14/16 87.5 Johansson et al. (Johansson et al., 2005) 1 0 0 0 0 1 0 0 1 0 NA NA NA NA NA 1 1 1 0 0 6/15 40.0 El-Sayed et al. (El-Sayed et al., 2014) 1 1 1 1 1 1 1 1 1 1 1 1 NA 0 NA 1 1 1 0 1 16/18 88.9 Wu et al. (Wu et al., 2011) 1 1 0 1 0 1 1 0 NA 0 NA NA NA NA NA NA NA 1 NA NA 6/10 60.0 Sup et al. (Sup et al., 2007b) 1 1 1 1 1 1 1 1 1 1 1 1 1 NA NA 1 1 1 0 1 17/18 94.4 Gholamreza et al. (Khademi et al., 2019) 1 1 1 1 1 1 1 1 1 1 1 1 NA NA NA NA NA 1 0 1 14/15 93.3 Huang et al. (Huang et al., 2011) 1 1 1 1 0 1 1 0 1 0 NA NA NA NA NA 1 1 1 NA 0 10/14 71.4 Jasni et al. (Jasni et al., 2019) 1 1 1 1 1 1 1 1 1 1 1 1 1 0 NA 1 1 1 0 1 17/19 89.5 Jasni et al. (Jasni et al., 2016) 1 1 0 1 1 1 1 1 1 1 1 1 NA 0 NA 1 1 1 0 1 17/18 94.4 Bhakta et al. (Bhakta et al., 2020) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Liu et al. (Liu et al., 2017) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Wu et al. (Wu et al., 2021) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Shaikh & Malhotra (Shaikh & Malhotra, 2020) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Wen et al. (Wen et al., 2020) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 Kadhim et al. (Kadhim et al., 2020) 1 1 0 1 1 1 1 1 1 1 NA NA NA NA NA 1 1 1 0 1 13/15 86.7 * Items related to external validity (items 11, 12, 13), internal validity (items 14, 15, 16, 17), and internal validity – confounding selection bias (items 21, 22, 23, 24, and 26) from the Downs and Black (D&B) criteria (D. a Winter, and a O, 1975) were not applicable in most studies for the present review, being applied only to evaluate randomized controlled trial reports. Abbreviations: D&B criteria met = 1, D&B criteria unmet = 0, Unable to determine = 0, Criteria not applicable to the study = NA. “Downs and Black” – Criteria are as summarized follows: 1-Hypothesis stated, 2- Outcome described in Introduction/ Method, 3- Participants’ characteristics described, 4- Intervention described, 5- Principal confounder in each group subjects described, 6- Findings described, 7- Data distribution reporting, 8- Description of adverse events, 9- Characteristics of patients lost to follow up described, 10- Exact p-values reported, 18- Statistical tests used to assess the main outcomes reported, 19- Adherence to the intervention described, 20- Accuracy of outcome measures described, 25- Adjustment for confounding in the analyses from which the main findings are reported. 4 Conclusion The control interface mechanisms for intention detection that may be used in an active prosthetic limb are realistically scientifically supported in this work. The quality of the studies varied from acceptable to good, with scores between 40% and 100% (Table 4 ), with a mean score of 83.3%, which is regarded as excellent. This is according to the accepted D&B checklist (AT, 2008 ). All studies accurately described the research's purpose and underlying premise (criteria 1) as well as the reliability of its outcome measures (criteria 20). Overall, the results of these investigations came from trustworthy lab tests. The paucity of application of the probability value results indicates that few of them provided clinically pertinent data. Most D&B criterion items demonstrate how few finds and advances are generally practical. The user's-biological-input focused sensory system in microprocessor active prosthetic legs is the best way to identify an amputee user's gait intention, it can be confidently inferred based on the quality outcome of this review. This study underlines and draws attention to many key and significant lines of research that have been presented throughout the years: Research lines examining several intention-detection sensors that can be used into active transfemoral prosthetic legs Research focuses on the different feedback and control techniques utilized for intention detection. Research areas focused on evaluating the effectiveness of active transfemoral prosthetic legs. The first study line explored the numerous sensory systems that, depending on the requirements needed by the researcher to construct the prosthetic leg, might be included into the active prosthetic leg for intention detection reasons. A prosthetic limb device's control architecture framework methodologies were covered in the second study area. Depending on the prosthesis functionality and the kinds of sensory systems used in the prosthetic legs, several control interface frameworks are established for intention detection. Given the available actuators for active prosthesis, weight-vs-size incompatibility has proven to be a design constraint. This highlights the necessity of creating lighter parts, such as titanium as its main material for prosthetic knee joint design or other comparable possibilities, or concentrating on the creation of artificial muscles. In order to limit muscular activity and make up for the energy users waste when using the prosthetic device, the usage of EMG sensors was highlighted. Nevertheless, despite the present improvements in prosthetic technology, further research is still required to develop more effective processing methods for recognizing and filtering biological signals. The last claim relates to the performance evaluation of a certain prosthesis to confirm the caliber of prosthetic devices. The majority of research employed motion lab analysis to collect certain metrics in real time, which were used as conclusion-making judgement criteria. Only commercially accessible technologies have been utilized for actual environmental evaluations. The ability to operate fitted prosthetic devices would be considerably improved if all three issues were addressed during technological development. The type of sensory system utilized in prosthetic legs affects the control algorithms employed in such legs. For instance, the mid-level controller often conducts the identification of the present state that the device is in while controlling the prosthetic device-oriented sensory system. Data from the prosthetic device itself will be collected by the sensors, which will then send it to the controller for interpretation before sending the algorithm to the actuator system. While this was going on, the user-biological-oriented sensory system and the neuro-mechanical fusion sensory system were both controlled by two levels of controllers. The high-level controller (fuzzy) would collect the signal from the sensory system (EMG or EEG) and analyze it before sending it to the low-level controller (PID) to instruct the actuator of the prosthetic leg system. Additionally, several methodologies might be employed to verify the effectiveness of each sensory system used in the prosthetic legs. One of these methods involves utilizing a motion analysis device to record every kinematic and kinetic parameter as amputees walked on prosthetic legs. To confirm the dependability of the sensors, the output signals from the sensory system will be mapped with the output from the motion analysis system. To assess the functioning of an active prosthetic leg's sensory system, the majority of research typically utilized five to 10 individuals. The tendency of correctness of the research study's sensory system increases with the number of people engaged. Further examination of the research shown in Tables 2 and 3 reveals that, according to the above-mentioned needed criteria, investigations involving the prosthetic-device oriented sensory system were more developed than those involving the other two options. A few investigations, including those on the CYBERLEGS and VI Knee reported using the produced active transfemoral prosthetic leg prototype, which was outfitted with and operated using the suggested sensory system. It demonstrates the researchers' faith in the suggested system's ability to keep the individuals safe. The user's-biological-input-oriented and neuro-mechanical fusion sensory system research is still in its early stages. It has been demonstrated that the impedance-based control method effectively coordinates the movements of the prosthetic device-oriented sensory system. Additionally, it was shown to be employed as an input signal to distinguish between different ambulation modes. However, the evaluation reveals that only one study reported using a kinematic sensor exclusively to carry out the job, whereas the others reported using both kinematic and kinetic sensors to detect activity mode transitions. However, as of now, this kind of sensory system's effectiveness is still insufficient to be employed in determining the user's purpose. This is due to the fact that there is no direct contact between the sensors and the user's body, and it is also unable to access the user's neurological input, which has been shown to be capable of deciphering motion intent before a movement really occurs. Additionally, the sensory system that is focused on prosthetic devices does not incorporate information about environmental responses. For example, an able-bodied person's movement pattern is also influenced by information from the sensory organs (such as the eyes and hearing) that can detect obstacles in the way. In order to avoid the obstruction, the brain will tell the muscles to flex their knees more and swing for a longer amount of time. The safety of the prosthesis user might be jeopardized in the absence of this information. In their study, Martin et al. asserted that if there is a discrepancy between the user's planned motion and the motion established by the prosthetic device controller, it may endanger the user's safety, cause gait deficit, and ultimately result in prosthetic device rejection. Using a sensory system that gathers data from the user's bodily input, such as EEG and EMG, is one technique to prevent this. EMG electrodes are used more often in lower-limb prosthetic devices. Even though several research showed that the EMG signal is reliable in determining the user's intention prior to the action actually occurring, its performance is constrained by a few limitations. It is extremely vulnerable to electrode-skin conductivity, motion artefacts, electrode misalignment, and intermuscular communication. Additionally, the EMG sensing system's don and doff procedures are laborious. Therefore, it can be projected that an EMG-only based sensory system will not be compelling enough to be employed in lower-limb MPC prosthetic until these signal-quality related concerns are properly solved. The controller receives more detailed information from the fusion sensory system. This sort of sensory system can measure muscle activation as well as kinetic and kinematic data, which enables the development of robust control algorithms and smooth switching between ambulation modes for MPC prostheses. In terms of processing time and RAM and ROM capacity of the processor, the complexity of the system to handle heavy information from the sensors (such as EMG and mechanical sensors) may limit the usefulness of the sensory system. According to research by Zhang et al., kinematics information was not as helpful as EMG and GRF/moment data for TF amputees doing real-time intent detection tasks. However, a more recent work by Stolyarov et al. shown that the intent recognition task could be accomplished using simply inertial sensors and the translational motion tracking approach. The sensory and a control system of a microprocessor-controlled prosthetic limb were reviewed in this literature study's conclusion. The classification of different sensory and control systems into new categories was put forth. In essence, depending on what researchers require in the prosthetic limb system, each form of sensor and its control has advantages and disadvantages. The sensory system designed for prosthetics may operate more effectively and would be useful for controlling how the device moves. Additionally, researchers working on prosthetic devices frequently employ this kind of sensory system, which is currently available on the market. This sensory system's shortcomings include a decreased ability to foretell users' intentions. Studies that employ user-biologically oriented input are effective in identifying users' intentions, but they lack dependability in real-world settings. One of the identified issues with using these kinds of sensors is that the output signal's resilience isn't compelling enough to be employed in active transfemoral prostheses. Fusion sensory systems are the finest choice of sensors for upcoming studies since they are reliable in giving the controller adequate data because they employ both mechanical and user-biological sensors. Additionally, when used in conjunction with an effective control system, these sensors might offer a smooth transition mode for active transfemoral prostheses. Therefore, our investigation revealed that the adoption of a simpler sensory system to operate an active transfemoral prosthetic limb is feasible with an inventive strategy to extrapolate the limited sensor information into a broader group of useful information. As a result, the two requirements for an effective sensory system—practicality and quality of the sensory data—can be met. Given the linked biomechanics of human gait, which the brain regulates subconsciously, it is regarded as a complicated activity. The difficulty in prosthetic technology still exists, which explains why no prosthetic device has been able to replace a human limb in the same way that the original one would function. Even with something as straightforward as walking, the difficulty comprises several elements that may be divided into various study areas. No matter how modest, this requires contributions from many fields of expertise. Abbreviations D&B Down and Black Checklist EMG Electromyography CC Classical Control ANFIS Adaptive Neuro-fuzzy Inference System Declarations Acknowledgment This research was partially supported by the University Malaya Faculty Research Grant (GPF068A-2018). The first author received a scholarship from the Department of Public Services, Malaysia (JPA Scholarship). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Author Contributions Nur Hidayah: Conceptualization, Methodology, Data curation, Data analysis, Writing – Original draft preparation. Nur Azah.: Data curation, Writing- review & editing, Supervision, Funding acquisition. Farahiyah: Data curation. Khin Wee: Data curation, Fanny Oddon – literature search and manuscript writing. Funding Geran Penyelidikan Fakulti - Faculty Research Grant code GPF068A-2018 and Jabatan Pertahanan Awam (JPA) Scholarships. Data Availability Statement No datasets were generated or analyzed during the current study. References Aeyels, B., Van Petegem, W., Sloten, J. V., Van Der Perre, G., Peeraer, L. (1995). An EMG-based finite state approach for a microcomputer-controlled above-knee prosthesis. Proc. 17th Int. Conf. Eng. Med. Biol. Soc. 2, 1315–1316. Alzaydi, A. A., Cheung, A., Joshi, N., Wong, S. (2011). Active prosthetic knee fuzzy logic-PID motion control, sensors and test platform design. Int. J. Sci. Eng. Res. 2, 1–17. Andrysek, J., Naumann, S., Cleghorn, W. L. (2005). Design and quantitative evaluation of a stance-phase controlled prosthetic knee joint for children. IEEE Trans. Neural Syst. Rehabil. Eng. 13, 437–443. doi: 10.1109/TNSRE.2005.856071. AT. (2008). Studies. Downs Black Checklist. Au, S., Berniker, M., Herr, H. (2008). Powered ankle-foot prosthesis to assist level-ground and stair-descent gaits. Neural Netw. 21, 654–666. doi: 10.1016/j.neunet.2008.03.006. Babuška, R., Verbruggen, H. B. (1996). An overview of fuzzy modeling for control. Control Eng. Pract. 4, 1593–1606. doi: 10.1016/0967-0661(96)00175-X. Bai, O., Kelly, Graham., Ding, Yu Fei., Douglas, Murphy., John, Fox., Brian, Burkhardt., et al. (2016). A wireless, smart EEG system for volitional control of lower-limb prosthesis. IEEE Reg. 10 Annu. Int. Conf. Proc. TENCON, 2016, 1–6. Bellmann, M., Schmalz, T., Blumentritt, S. (2010). Comparative biomechanical analysis of current microprocessor-controlled prosthetic knee joints. Arch. Phys. Med. Rehabil. 91, 644–652. doi: 10.1016/j.apmr.2009.12.014. Bhakta, K., Camargo, J., Kunapuli, P., Childers, L., Young, A. (2020). Impedance control strategies for enhancing sloped and level walking capabilities for individuals with transfemoral amputation using a powered multi-joint prosthesis. Mil. Med. 185, 490–499. doi: 10.1093/milmed/usz229. Borjian, R., Lim, J., Khamesee, M. B., Melek, W. (2008). The design of an intelligent mechanical active prosthetic knee. IECON 34th Annu. Conf. Ieee Ind. Electron. Soc. Vols 1, 5, Proc. 2918–2923. Brantley, J. A., Luu, T. P., Nakagome, S., Contreras-Vidal, J. L. (2017). Prediction of lower-limb joint kinematics from surface EMG during overground locomotion IEEE Int. Conf. Syst. Man, Cybern. SMC 2017, 1705–1709. Celebi, B., Yalcin, M., Patoglu, V. (2013). AssistOn-Knee: A self-aligning knee exoskeleton, in IEEE Int. Conf. Intell. Robots Syst. 996–1002. Dawley, J. A., Fite, K. B., Fulk, G. D. (2013). EMG control of a bionic knee prosthesis: Exploiting muscle co-contractions for improved locomotor function. IEEE Int. Conf. Rehabil. Robot.. 2013, 6650389. doi: 10.1109/ICORR.2013.6650389. Delis, A. L., Carvalho, J. L. A., Da Rocha, A. F., Ferreira, R. U., Rodrigues, S. S., Borges, G. A. (2009). Estimation of the knee joint angle from surface electromyographic signals for active control of leg prostheses. Physiol. Meas. 30, 931–946. doi: 10.1088/0967-3334/30/9/005. Downs, S. H., Black, N. (1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J. Epidemiol. Community Health. 52, 377–384. doi: 10.1136/jech.52.6.377. Dyck, W. R., Onyshko, S., Hobson, D. A., Winter, D. A., Quanbury, A. O.W. R. Dyck, S. Onyshko, D., D. a Winter, and a O. (1975). Bull. Prosthet. Res. 169–186. El-Sayed, A. M., Hamzaid, N. A., Abu Osman, N. A. A. (2014). Piezoelectric bimorphs “characteristics as in-socket sensors for transfemoral amputees”. Sensors. 14, 23724–23741. doi: 10.3390/s141223724. El-Sayed, A. M., Hamzaid, N. A., Azuan, N., Osman, A. (2014). Applied mathematics& information sciences modelling and control of a linear actuated transfemoral knee joint in basic daily movements. Appl. Math. Inf. Sci. 7, 1–11. Fite, K., Mitchell, J., Sup, F., Goldfarb, M. (2007). Design and control of an electrically powered knee prosthesis, in 10th International Conference on Rehabilitation Robotics, ICORR’07 (IEEE Publications). Flowers, W. C. (1971). ‘A man-interactive simulator system for an above-knee prosthetics studies,’ Massachusettes Institute of Technology. Furuya, Y., Mikami, T., Suzuki, T., Fujimoto, Y. (2013). On an active prosthetic knee joint driven by a high thrust force helical motor. IECON Proc. (Ind. Electron. Conf.). 2, 5894–5899. Grimmer, M., Eslamy, M., Seyfarth, A. (2014). Energetic and peak power advantages of series elastic actuators in an actuated prosthetic leg for walking and running. Actuators. 3, 1–19. doi: 10.3390/act3010001. Ha, K. H., Varol, H. A., Goldfarb, M. (2011). Volitional control of a prosthetic knee using surface electromyography. IEEE Trans. Biomed. Eng. 58, 144–151. doi: 10.1109/TBME.2010.2070840. Hafner, B. J., Askew, R. L. (2015). Physical performance and self-report outcomes associated with use of passive, adaptive, and active prosthetic knees in persons with unilateral, transfemoral amputation: Randomized crossover trial. J. Rehabil. Res. Dev. 52, 677–700. doi: 10.1682/JRRD.2014.09.0210. Hafner, B. J., Willingham, L. L., Buell, N. C., Allyn, K. J., Smith, D. G. (2007). Evaluation of function, performance, and preference as transfemoral amputees transition from mechanical to microprocessor control of the prosthetic knee. Arch. Phys. Med. Rehabil. 88, 207–217. doi: 10.1016/j.apmr.2006.10.030. Hamzaid, N. A., Mohd Yusof, N. H., Jasni, F. (2020). Sensory systems in micro-processor controlled prosthetic leg: A review. IEEE Sensors J. 20, 4544–4554. doi: 10.1109/JSEN.2019.2944653. Hargrove, L. J., Simon, A. M., Lipschutz, R., Finucane, S. B., Kuiken, T. A. (2013). Non-weight-bearing neural control of a powered transfemoral prosthesis. J. Neuroeng. Rehabil. 10, 62. doi: 10.1186/1743-0003-10-62. Hargrove, L. J., Simon, A. M., Young, A. J., Lipschutz, R. D., Finucane, S. B., Smith, D. G. et al. (2013). Robotic leg control with EMG decoding in an amputee with nerve transfers. N. Engl. J. Med. 369, 1237–1242. doi: 10.1056/NEJMoa1300126. Huang, H., Zhang, F., Hargrove, L. J., Dou, Z., Rogers, D. R., Englehart, K. B. (2011). Continuous locomotion-mode identification for prosthetic legs based on neuromuscular–mechanical fusion. IEEE Trans. Biomed. Eng. 58, 2867–2875. doi: 10.1109/TBME.2011.2161671. Huang, H., Zhang, F., Sun, Y. L., He, H. (2010). Design of a robust EMG sensing interface for pattern classification. J. Neural Eng. 7, 056005. doi: 10.1088/1741-2560/7/5/056005. Jasni, F., Hamzaid, N. A., Al-Nusairi, T. Y., Mohd Yusof, N. H. Mohd., Shasmin, H. N., Ng, S. C. (2019). Feasibility of a gait phase identification tool for transfemoral amputees using piezoelectric- based in-socket sensory system. IEEE Sensors J. 19, 6437–6444. doi: 10.1109/JSEN.2019.2910161. Jasni, F., Hamzaid, N. A., Muthalif, A. G. A., Zakaria, Z., Shasmin, H. N., Ng, S. (2016). In-socket sensory system for transfemoral amputees using piezoelectric sensors: An efficacy study. IEEE/ASME Trans. Mechatron. 21, 2466–2476. doi: 10.1109/TMECH.2016.2578679. Jasni, F., Hamzaid, N. A., T. Y. Al-nusairi, N. Hidayah, and M. Yusof. Feasibility of a gait phase identification tool for transfemoral amputees using piezoelectric- based in-socket sensory system. (2019). Johansson, J. L., Sherrill, D. M., Riley, P. O., Bonato, P., Herr, H. (2005). A clinical comparison of variable-damping and mechanically passive prosthetic knee devices. Am. J. Phys. Med. Rehabil. 84, 563–575. doi: 10.1097/01.phm.0000174665.74933.0b. Kadhim, D. A., Raheema, M. N., Hussein, J. S. (2020). Design of an Intelligent Controller for above Knee Prostheses based on an Adaptive Neuro-Fuzzy Inference System. IOP Conf. Ser.: Mater. Sci. Eng. 671. doi: 10.1088/1757-899X/671/1/012066. Khademi, G., Mohammadi, H., Simon, D. (2019). Gradient-based multi-objective feature selection for gait mode recognition of transfemoral amputees. Sensors (Basel). 19. doi: 10.3390/s19020253. Kirker, S., Keymer, S., Talbot, J., Lachmann, S. (1996). An assessment of the intelligent knee prosthesis. Clin. Rehabil. 10, 267–273. doi: 10.1177/026921559601000314. Liu, M., Zhang, F., Huang, H. H. (2017). An adaptive classification strategy for reliable locomotion mode recognition. Sensors (Basel). 17. doi: 10.3390/s17092020. Martinez-Villalpando, E. C., Herr, H. (2009). Agonist-antagonist active knee prosthesis: A preliminary study in level-ground walking. J. Rehabil. Res. Dev. 46, 361–373. Mohd Yusof, N. H. Mohd., Hamzaid, N. A., Jasni, F., Lai, K. W. (2018). In-socket sensory system with an adaptive neuro-based fuzzy inference system for active transfemoral prosthetic legs. J. Electron. Imag. 28, 1. doi: 10.1117/1.JEI.28.2.021002. Pfeifer, S., Pagel, A., Riener, R., Vallery, H. (2015). Actuator with angle-dependent elasticity for biomimetic transfemoral prostheses. IEEE/ASME Trans. Mechatron. 20, 1384–1394. doi: 10.1109/TMECH.2014.2337514. Pieringer, D. S., Grimmer, M., Russold, M. F., Riener, R. (2017). Review of the actuators of active knee prostheses and their target design outputs for activities of daily living. IEEE Int. Conf. Rehabil. Robot.. 2017, 1246–1253. doi: 10.1109/ICORR.2017.8009420. Saglam, Y., Gulenc, B., Birisik, F., Ersen, A., Yilmaz Yalcinkaya, E., Yazicioglu, O. (2017). The quality-of-life analysis of knee prosthesis with complete microprocessor control in trans-femoral amputees. Acta Orthop. Traumatol. Turc. 51, 466–469. doi: 10.1016/j.aott.2017.10.009. Segal, A. D., Orendurff, M. S., Klute, G. K., McDowell, M. L., Pecoraro, J. A., Shofer, J. et al. (2006). Kinematic and kinetic comparisons of transfemoral amputee gait using C-Leg and Mauch SNS prosthetic knees. J. Rehabil. Res. Dev. 43, 857–870. doi: 10.1682/jrrd.2005.09.0147. Shaikh, S., Malhotra, A. (2020). Real-time feedback control for knee prosthesis using motion fusion algorithm in 6-dof IMU. J. Sci. Ind. Res. (India). 79, 213–215. Song, L., Wang, X., Gong, S., Shi, Z., Chen, L. (2008). Design of active artificial knee joint. IFMBE Proc. 19, 155–158. doi: 10.1007/978-3-540-79039-6_40. Sup, F., Bohara, A., Goldfarb, M. (2007a). Design and control of a powered knee and ankle prosthesis, in Proc. - IEEE International Conference on Robotics and Automation, 4134–4139. Sup, F., Bohara, A., Goldfarb, M. (2007b). Design and control of a powered knee and ankle prosthesis. Proc. – IEEE Int. Conf. Robot. Autom., no. April, 4134–4139. Sup, F., Varol, H. A., Mitchell, J., Withrow, T., Goldfarb, M. (2008a). Design and control of an active electrical knee and ankle prosthesis. Proc. 2nd Bienn. IEEE/RAS-EMBS Int. Conf. Biomed. Robot. Biomechatronics, Scottsdale, AZ, USA, 523–528. Sup, F., Varol, H. A., Mitchell, J., Withrow, T., Goldfarb, M. (2008b). Design and control of an active electrical knee and ankle prosthesis, in Proc. 2nd Biennial IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics. Proc. IEEE RAS EMBS Int. Conf. Biomed. Robot. Biomechatron. 2008, 523–528. doi: 10.1109/BIOROB.2008.4762811. Thatte, N., Geyer, H. (2016). Toward balance recovery with leg prostheses using neuromuscular model control. IEEE Trans. Biomed. Eng. 63, 904–913. doi: 10.1109/TBME.2015.2472533. Thiele, J., Westebbe, B., Bellmann, M., Kraft, M. (2014). Designs and performance of microprocessor-controlled knee joints. Biomed. Tech. 59, 65–77. Torrealba, R. R., Fernández-López, G., Grieco, J. C. (2008). Towards the development of knee prostheses: Review of current researches. Kybernetes. 37, 1561–1576. Varol, H. A., Goldfarb, M. (2007a). Decomposition-based control for a powered knee and ankle transfemoral prosthesis 10th Int. Conf. Rehabil. Robot. ICORR’07, 00 (IEEE Publications), 783–789. Varol, H. A., Goldfarb, M. (2007b). Decomposition-based control for a powered knee and ankle transfemoral prosthesis 10th Int. Conf. Rehabil. Robot. ICORR’07, 00, no. c, (IEEE Publications), 783–789. Varol, H. A., Sup, F., Goldfarb, M. (2009). Real-time gait mode intent recognition of a powered knee and ankle prosthesis for standing and walking. Proc. IEEE RAS EMBS Int. Conf. Biomed. Robot Biomechatron. 2008, 66–72. doi: 10.1109/BIOROB.2008.4762860. Varol, H. A., Sup, F., Goldfarb, M. (2010). Multiclass real-time intent recognition of a powered lower limb prosthesis. IEEE Trans. Biomed. Eng. 57, 542–551. doi: 10.1109/TBME.2009.2034734. Wen, Y., Si, J., Brandt, A., Gao, X., Huang, H. H. (2020). Online reinforcement learning control for the personalization of a robotic knee prosthesis. IEEE Trans. Cybern. 50, 2346–2356. doi: 10.1109/TCYB.2019.2890974. Wu, R., Li, M., Yao, Z., Si, J., He, Huang. (2021). Reinforcement learning enabled automatic impedance control of a robotic knee prosthesis to mimic the intact knee motion in a co-adapting environment. Wu, S. K., Waycaster, G., Shen, X. (2011). Electromyography-based control of active above-knee prostheses. Control Eng. Pract. 19, 875–882. doi: 10.1016/j.conengprac.2011.04.017. Young, A. J., Kuiken, T. A., Hargrove, L. J. (2014). Analysis of using EMG and mechanical sensors to enhance intent recognition in powered lower limb prostheses. J. Neural Eng. 11, 056021. doi: 10.1088/1741-2560/11/5/056021. Young, A. J., Simon, A. M., Hargrove, L. J. (2014). A training method for locomotion mode prediction using powered lower limb prostheses. IEEE Trans. Neural Syst. Rehabil. Eng. 22, 671–677. doi: 10.1109/TNSRE.2013.2285101. Young, A. J., Simon, A., Hargrove, L. J. (2013). An intent recognition strategy for transfemoral amputee ambulation across different locomotion modes. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS. 2013, 1587–1590. doi: 10.1109/EMBC.2013.6609818. Zlatnik, D. (1998). Intelligently controlled above knee (A/K) prosthesis Inst. Robot. ETH- Zurich. Switz. Cite Share Download PDF Status: Published Journal Publication published 31 Oct, 2025 Read the published version in Physical and Engineering Sciences in Medicine → Version 1 posted Editorial decision: Major revisions 23 Nov, 2023 Reviewers agreed at journal 12 Jun, 2023 Reviewers invited by journal 18 Apr, 2023 Editor invited by journal 16 Apr, 2023 Editor assigned by journal 15 Apr, 2023 First submitted to journal 13 Apr, 2023 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2814842","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":192997069,"identity":"44a20026-8f64-4313-8c91-22fef08b809b","order_by":0,"name":"nur hidayah mohd yusof","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4ElEQVRIiWNgGAWjYDCCAwwMzAwVEnIGYJ6BBbFaztgYGwApoBYJIrUwtqUlbgBrYSBCC9/xM4afC9sOp29n7z+64UeBBAN/e3cCXi2SZ3KMpWecO5y7s+cw280eoMMkzpzdgFeLwYEcA2messO5G24ks93gAWoxkMgloOX8G+PfPGyH0w2AWm7+IUrLjRwzaZ62tASQlttE2SJ541mZNc8ZG8MNZw6b3ZYxkOAh6Be+88mbb/NUSMgbHG98dvPNHxs5/vZe/FoYGDgMULg8BJSDAPsDIhSNglEwCkbBiAYAXbBKjBSwwhkAAAAASUVORK5CYII=","orcid":"","institution":"University of Malaya - City Campus","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"nur","middleName":"hidayah mohd","lastName":"yusof","suffix":""},{"id":192997070,"identity":"73db1a8e-89d3-4708-8559-24231cd3e822","order_by":1,"name":"Nur Hidayah Mohd Yusof","email":"","orcid":"","institution":"Universiti Malaya","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Nur","middleName":"Hidayah Mohd","lastName":"Yusof","suffix":""},{"id":192997071,"identity":"157a233f-bbd8-4a2d-9fb3-1f07a4e13ae8","order_by":2,"name":"Nur Azah Hamzaid","email":"","orcid":"","institution":"University of Malaya: Universiti Malaya","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Nur","middleName":"Azah","lastName":"Hamzaid","suffix":""},{"id":192997072,"identity":"36488e69-909a-4f60-98ed-099a474e8192","order_by":3,"name":"Lai Khin Wee","email":"","orcid":"","institution":"University of Malaya: Universiti Malaya","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lai","middleName":"Khin","lastName":"Wee","suffix":""},{"id":192997073,"identity":"2ca7b462-ced3-430d-9c58-3faa4471f201","order_by":4,"name":"Farahiyah Jasni","email":"","orcid":"","institution":"International Islamic University Malaysia","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Farahiyah","middleName":"","lastName":"Jasni","suffix":""},{"id":192997074,"identity":"cfe8c592-893c-44b9-86dd-80a553cf65b0","order_by":5,"name":"Fanny Oddon","email":"","orcid":"","institution":"Polytech Montpellier","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Fanny","middleName":"","lastName":"Oddon","suffix":""}],"badges":[],"createdAt":"2023-04-14 00:59:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2814842/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2814842/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s13246-025-01623-0","type":"published","date":"2025-10-31T15:58:39+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":36064960,"identity":"f52a0521-30da-4917-894e-2b946637e4dd","added_by":"auto","created_at":"2023-04-20 13:36:18","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":588570,"visible":true,"origin":"","legend":"\u003cp\u003ePrisma flowchart for included and excluded studies in the systematic review on intention detection recognition for an active prosthetic leg.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-2814842/v1/4262bb1ee3f4d6e0d1dd2ae3.jpeg"},{"id":36064961,"identity":"557741a5-023a-4dda-81e9-8868d97d2286","added_by":"auto","created_at":"2023-04-20 13:36:18","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":178440,"visible":true,"origin":"","legend":"\u003cp\u003eSystem configuration and architecture of controller system for an active prosthetic leg.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-2814842/v1/3546f11318bf58e7ece5e0b2.jpeg"},{"id":95040032,"identity":"802ccee8-4e53-46cb-87c0-13ed6d8731db","added_by":"auto","created_at":"2025-11-03 16:07:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2732738,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2814842/v1/6e9125f0-5a2f-4f94-93eb-87765ac4bc2e.pdf"}],"financialInterests":"","formattedTitle":"Control Interfaces for Intention Detection in Active Transfemoral Prosthetics: A Systematic Review","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eOver the past 10 years, several electrical and sensory technologies have been included into prosthetic knee joint systems (Hargrove et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), mostly to create the best prosthesis that mimics the typical gait of a person. Three categories can be used to categorise contemporary transfemoral prosthetic devices: passive knee joints (Hafner and Askew, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Celebi et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Johansson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2005\u003c/span\u003e); mechanical knee joints (Johansson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Borjian et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Varol and Goldfarb, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2007a\u003c/span\u003e; Hargrove et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e); and active knee joints, either powered or (Segal et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; D. A Winter et al, and Ao et al, 1975; Song et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Martinez-Villalpando and Herr, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). The market now offers only energetically passive transfemoral prosthetic devices that are commercially available. To offer enough ground clearance, the majority of passive device mechanisms restrict knee flexion during the stance phase and execute it during the swing phase. To prevent the knee from buckling and dropping, the user must fully extend their knee upon heel impact in order to lock the joint. Amputees must meet these two requirements in order to completely control a passive prosthetic knee joint. The knee flexion and extension angular velocities, however, may be changed during swinging by using a damper. Amputees can effortlessly manipulate their prosthetic legs thanks to this technology.\u003c/p\u003e \u003cp\u003eThrough the development of innovative solutions that help patients restore their regular stride, several studies have been conducted throughout the years to advance prosthetic technology. Using a microprocessor as a controller is one approach (Bellmann et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Thiele et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). There are two categories of microprocessor-controlled prosthetic legs: mechanically passive and mechanically active (Hafner and Askew, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Segal et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Hafner et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Sup et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2008a\u003c/span\u003e). Both of these technologies may bring back a number of locomotor modes that analyse and interpret information from the prosthetic legs' sensory system using finite state machines, such as mechanically oriented sensors or physiologically oriented sensors (Hamzaid et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). There is currently a prosthesis available that can assess state changes in locomotor modes (Fite et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Au et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). But to overcome the limitations of microprocessor-controlled prostheses, an intent detection approach was created to allow seamless, automatic, and natural transitions between locomotor modes (Young et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). This technique, as well as others, may be used to mechanically active or passive microprocessor-controlled prostheses.\u003c/p\u003e \u003cp\u003eIn light of this recent development, the goal of this study is to illustrate the various scientific research avenues that have been explored during the previous 10 years, with considerations on their experimental methods and findings through D\u0026amp;B analysis. Earlier research (Babuka and Verbruggen, 1996; Torrealba et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Pieringer et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) reviewed the designs of various knee prosthesis, but this was performed by identifying the intent detection technique and classifying its control interface and approach. This study emphasises the key elements of the prosthetic leg control literature: sensors for intention detection, control strategies, and performance assessment methods. In each paragraph, the components or pieces needed to build a sophisticated knee prosthetic system are highlighted. Finally, a description of the evolution of prosthetic technology systems from the simplest to the most complex is given.\u003c/p\u003e"},{"header":"2 Methodology","content":"\u003ch2\u003e2.1 Literature Search Strategy\u003c/h2\u003e\n\u003cp\u003eWithin the previous 18 years, from May 2005 to June 2022, a literature search was carried out utilising the Google Scholar, Web of Science, IEEE Xplore, and PubMed databases. Based on the framework\u0026apos;s investigation into intention detection, the year was selected. The phrases \u0026quot;sensory system,\u0026quot; \u0026quot;controller,\u0026quot; \u0026quot;gait intention recognition,\u0026quot; and \u0026quot;microprocessor control\u0026quot; in combination with \u0026quot;active prosthetic transfemoral leg\u0026quot; were used to extract pertinent search terms, keywords, or publications. Studies that were reported in English and satisfied the qualifying requirements were kept for additional investigation.\u003c/p\u003e\n\u003ch2\u003e2.2 Eligibility Criteria\u003c/h2\u003e\n\u003cp\u003eThis review concentrated on transfemoral amputees\u0026apos; usage of prosthetic devices. The investigations were not just proof-of-concept research in the field, but also studies that reported prototypes of active prosthetic legs. Additionally, if the system was developed and tested on orthoses or upper limb prostheses or if the criteria had the potential to be applied to prosthetic devices generally, the \u0026quot;intention detection\u0026quot; criteria were also taken into account in this evaluation. Due to poor data presentation and result validation, conference papers and proceedings were omitted from this review.\u003c/p\u003e\n\u003ch2\u003e2.3 Data Extraction from Selected Articles\u003c/h2\u003e\n\u003cp\u003eFour writers conducted this review (NH, NAH, LKW, and FJ). The titles and abstracts of the research were used to find those that could be eligible. We only kept and studied full-length papers that described the active control mechanism and user intention detection. Studies were included if the user intention and its sensory system were applied to prosthetics and orthotics as the primary component of the reported creation of the intervention of interest. Passive prosthetic devices were not included in the studies for this evaluation since the co-interventions might have influenced the findings or scope of the analysis.\u003c/p\u003e\n\u003ch2\u003e2.4 Assessment of Study Quality\u003c/h2\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003eThe Downs and Black checklist were created to assess the methodological excellence of comparison research, both randomized and nonrandomized. The checklist\u0026apos;s 27 elements include reporting, external validity, internal validity (bias and confounding), and power, which are all methodological components. One item was graded on a 3-point scale (yes\u0026thinsp;=\u0026thinsp;2, partial\u0026thinsp;=\u0026thinsp;1, no\u0026thinsp;=\u0026thinsp;0), while the other 26 items were assessed as either yes (=\u0026thinsp;1) or no/unable to identify (=\u0026thinsp;0). Values range from 0 to 28, and higher scores indicate that the study\u0026apos;s methodology was sounder. To classify research by quality, the following cut-points have been proposed: excellent (26\u0026ndash;28), good (20\u0026ndash;25), fair (15\u0026ndash;19), and bad (14). The Downs and Black checklist have been shown to have adequate psychometric qualities, including internal consistency, test-retest reliability, inter-rater reliability, and criterion validity, in other publications. The top six quality evaluation techniques eligible for use in systematic reviews include the checklist. The Downs and Black (D\u0026amp;B) checklist for quality assessment\u0026apos;s search approach was used to perform this systematic literature review (Downs and Black, \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e). The search strategy specifies the query, databases, and inclusion and exclusion standards for sorting the records that are obtained. The control interface tactics used by an active prosthetic limb and its sensory system are the major subject of this review.\u003c/p\u003e\n\u003cp\u003eThe D\u0026amp;B (AT, \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e) checklist methodological criteria were used to evaluate the review studies\u0026apos; quality. Some of the criteria were disregarded since they applied solely to randomized controlled trials and were not relevant to the current development of prostheses. Examples of items that were not applicable in this review were those that dealt with external validity (items 12 and 13), internal validity (items 14, 15, and 16), and internal validity - confounding selection bias (items 21, 22, 23, and 24). In order to compare the methodological quality of the included studies, the D\u0026amp;B score was normalized based on the grading scale of the inclusion and exclusion criteria of the study.\u003c/p\u003e\n\u003ch2\u003e2.5 Classification Table\u003c/h2\u003e\n\u003cp\u003eTo obtain an overview of all available prosthetic knee devices, the required criteria or components are summarized in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. The main characteristics of the different systems were structured according to the following criteria:\u003c/p\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSummary of control frameworks strategies for an active prosthetic leg\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eControl strategy\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eControl modes\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eActuation principles\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSensory systems\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eClassical control (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Furuya et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eClassical\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eElectro-mechanical\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMechanical\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMachine learning technique (Bellmann et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Alzaydi et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e; Kadhim et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntelligent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePneumatic \u0026amp; powered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMechanical\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePattern recognition (Young et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Huang et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e; Brantley et al., \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e; Shaikh and Malhotra, \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntelligent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHydraulic \u0026amp; powered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMechanical \u0026amp; biological\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVolitional control (Dawley et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e; Bai et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e\u0026ndash;; Ha et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntelligent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHydraulic \u0026amp; powered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMechanical \u0026amp; biological\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMyoelectric control (Young et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Aeyels et al., \u003cspan class=\"CitationRef\"\u003e1995\u003c/span\u003e; Delis et al., \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e; Huang et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Wu et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntelligent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHydraulic \u0026amp; powered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBiological\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEcho control (Varol and Goldfarb, \u003cspan class=\"CitationRef\"\u003e2007b\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntelligent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHydraulic \u0026amp; powered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBiological\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u0026nbsp;Prosthetic knee joint: This acts as an identifier for the prosthetic knee joint system.\u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eSensory system and signal: This indicates the types of sensors used in the transfemoral prosthetic leg and the signal acquired.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eActuator characteristics: This indicates the characteristics of the actuator used in the prosthetic leg concerning the motor, transmission, power rating of the actuator, and moment rating at the knee joint.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eControl strategies: This indicates the control strategies used in the prosthetic leg to connect both sensory and actuator systems so that seamless transition of locomotion modes can be achieved.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eStage: This indicates whether the prototype is in the lab experiment stage, simulation, platform test design, or is already commercialized in the market.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003ePerformance assessment tools: Indicates the tools or equipment used in the studies to assess or analyze the performance of the active prosthetic leg.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAmbulation tasks: This indicates whether the prosthetic knee device can perform all ambulation functions, such as level ground walking, stairs and ramp negotiation, and sit to stand.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eUser intention detection, which indicates the intention detection ability of an active transfemoral prosthetic leg.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eNumber of subjects: This is the number of subjects that participated in the study.\u003c/p\u003e\n \u003c/li\u003e\n\u003c/ul\u003e"},{"header":"3 Results And Discussion","content":"\u003ch2\u003e3.1 Characteristics of Included Studies\u003c/h2\u003e\n\u003cp\u003eInitial selections for inclusion included 17495 articles (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). After eliminating any duplicate or related papers following title and abstract screening, only 211 studies were retrieved and examined. After a thorough evaluation, 179 articles were eliminated since the inclusion requirements weren\u0026apos;t met. Last but not least, this review included, evaluated, and examined 39 papers (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). An active prosthetic leg\u0026apos;s control strategy framework and goal output were examined in fifteen (15) trials. In two (2) trials, conventional control methods for transfemoral prosthetic legs were examined. Eight further research looked at the potential implementation of intent detection in the transfemoral prosthetic leg, while fourteen (14) papers explored the active prosthetic leg\u0026apos;s machine learning algorithm.\u003c/p\u003e\n\u003ch2\u003e3.2 Methodology Quality\u003c/h2\u003e\n\u003cp\u003eThe control interface mechanisms for intention detection that may be used in an active prosthetic limb are realistically scientifically supported in this work. The quality of the studies varied from acceptable to good, with scores between 40 and 100% (Table \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e), with a mean score of 83.3%, which is regarded as excellent. This is according to the accepted D\u0026amp;B checklist (AT, \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). All studies accurately described the research\u0026apos;s purpose and underlying premise (criteria 1) as well as the reliability of its outcome measures (criteria 20). Overall, the results of these investigations came from trustworthy lab tests. The paucity of application of the probability value results indicates that few of them provided clinically pertinent data. Most D\u0026amp;B criterion items demonstrate how few finds and advances are generally practical.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;The papers under consideration provided evidence with acceptable methodological quality that other researchers may use as a guide or a point of reference. The majority of these studies, with the exception of Dawley et al. (Dawley et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e) and Johansson et al., mentioned the result measured in their methods section (criterion 2) (Johansson et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e). All authors with the exception of Johansson et al. clearly explained the study\u0026apos;s intervention (criterion 4) (Johansson et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e). Additionally, every study\u0026mdash;aside from Dawley et al. (Dawley et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e\u0026ndash;clearly explained the study\u0026apos;s primary results) (criteria 6). With the exception of Varol et al, all studies adjusted for confounding in the analyses from which the main conclusions were mostly not presented (criterion 25) (Sup et al., 2007).\u003c/p\u003e\n\u003ch2\u003e3.3 Sensors in Transfemoral Prosthesis for Intention Detection\u003c/h2\u003e\n\u003cp\u003eThroughout the years, several questions have been raised to develop a good intent recognition system that can provide the best user intention recognition with fast and reliable predicted performance. Previous studies have addressed these questions in different ways. The mechanism of the sensory system can be addressed using a wide range of approaches and varying degrees of invasiveness. Generally, two types of sensory systems have been adopted for user intention detection: prosthetic device-oriented (Sup et al., \u003cspan class=\"CitationRef\"\u003e2008b\u003c/span\u003e; Sup et al., \u003cspan class=\"CitationRef\"\u003e2007a\u003c/span\u003e; Varol et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Young et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Yusof et al., 2018) and biological-oriented (Furuya et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e; Alzaydi et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e; Bai et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eProsthetic devices with user-oriented sensors primarily use mechanical sensors to detect the biomechanics of gait, such as the measurement of forces, torques, joint angles, and vertical orientation. For example, Sup and team (Sup et al., \u003cspan class=\"CitationRef\"\u003e2008b\u003c/span\u003e) used a load cell placed between the prosthesis and the user to derive joint torque and motion. The sensors measured the interaction forces and moments between the prosthesis and the user for control and intent recognition purposes. In addition, Sup et al. (Sup et al., 2007) used mechanical sensors mounted on a powered prosthesis as a control algorithm to predict user intention.\u003c/p\u003e\n\u003cp\u003eOn the other hand, Jasni and colleagues (Jasni et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; Jasni et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e) used piezoelectric sensors as in-socket sensory systems for intention detectors in active transfemoral prosthetic legs. The studies reported the efficacy of piezoelectric sensors to be used as a potential input signal for the control system of an active prosthetic leg using force profiles of muscle pressure on the socket and the corresponding ground reaction force while performing movements. In addition, El-Sayed and his team (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Mohd Yusof et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e) utilized piezoelectric transducers as a feedback signal source in the simulated control of an active prosthetic leg. Researchers have used movement characterizations based on sensor sensitivity against the stump, and thus provided an accurate voltage signal to the control feedback system. Although these studies have demonstrated the reliability of piezoelectric sensors for use as an in-socket sensory system for user movement and intention detection, they have yet to verify that such a signal can be used as an input to a feedback control system.\u003c/p\u003e\n\u003cp\u003eTo overcome the shortcomings of prosthetic device-oriented sensors, user-biological-oriented sensors have also been adopted. Surface EMG sensors are one of the major neural-control sources for prosthetic leg development that provide information for intention detection purposes from the muscle contractions in the amputee\u0026rsquo;s residual limb, and then give commands and modulate the output of the prosthesis.\u003c/p\u003e\n\u003ch2\u003e3.4 Control Strategies for Actuation System of an Active Prosthetic Leg\u003c/h2\u003e\n\u003cp\u003eThe purpose of having a control system is to execute the intended movement of the prosthetic knee and ensure a seamless transition mode during locomotion activity. The control system architecture of an active transfemoral prosthetic leg can be described as a three-level hierarchical controller scheme (Khademi et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e) as illustrated in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e. The high-level controller acts as an intent recognizer, and an intelligent control scheme was used at this level, consisting of three sub-parts: locomotion mode detector, slope calculator, and cadence calculator.\u003c/p\u003e\n\u003cp\u003eThe locomotion mode detector differentiates the ambulation modes of the user and sends the information to the middle-level controller, whereas booth calculators estimate the required amount of slope and cadence during walking.\u003c/p\u003e\n\u003cp\u003eThe middle-level controller controls the ambulation function mode (e.g., walking, slope, sit-to-stand, and climbing stairs) of the system, whereby it generates the required torque reference for the joints by modulating the variable knee impedance depending on the phase of the ambulation mode using a finite state machine. This intelligent control system monitors the movements and inputs of amputees and generates commands that the prosthesis actuator should perform. At the lowest level of the control hierarchy, a classical control (CC) scheme was used for transmission dynamics or feedback purposes. The intelligent control system is normally used in high-level controllers to ensure a reliable interaction with amputee users as well as to provide continuous transition between locomotion patterns.\u003c/p\u003e\n\u003ch2\u003e3.5 Low Level Controller (Classical)\u003c/h2\u003e\n\u003cp\u003eVarious CC schemes can be used as a controller system architecture for active prosthetic legs, depending on the type of prosthetic leg to be developed. The common method is a proportional integral derivative closed- and open-loop control. Proportional integral derivative control is a way of driving a system towards a target position or level, using a control loop feedback mechanism to control process variables. Normally, the CC strategy (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) is used to model the dynamic behavior of the prosthetic system itself before developing the actual one, and this strategy is used in a low-level controller. Researchers used this control method to construct a mathematical model of the system before implementing the actual model. Thus, the safety of the prosthetic leg is ensured. It is recommended to be used as a first-stage process to determine the kinematics and kinetic behavior of lower-limb prostheses. Classical control CC can be performed using MATLAB or any other simulation software, which is based on dynamic-mathematical modeling and is considered extremely complex but less robust. Thus, this type of control system is difficult to implement in real-time (Zlatnik, \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e). El-Sayed et al. (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) used this strategy to analyze the effectiveness of the actuation system of prosthetic legs, and the results presented a realistic simulation in terms of knee parameters. However, its real-time performance requires further verification.\u003c/p\u003e\n\u003ch2\u003e3.6 Middle Level Controller\u003c/h2\u003e\n\u003cp\u003eIn the middle-level controller, finite-state impedance control is used as a soft control scheme to emulate the impedance behavior of the healthy biomechanical gait. This control strategy was based on pre-stated rules that combine the information of certain knee parameters, thus allowing the control action of the prosthetic actuator system. This impedance strategy is emulated by modulating the impedance of the prosthetic knee joint according to the phase of the gait. Each gait phase had parameters for stiffness, damping, and spring equilibrium angle for the knee joint. In addition, the purpose of this finite state control strategy is to enable interaction between the user and prosthesis. That is, in any given state, the behavior is passive and would come to rest at local equilibrium, thus providing reliable and predictable behavior for the user. Zlatnik (Zlatnik, \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e) used this strategy as the knowledge base for a system that stored the predefined rules and parameters required to facilitate the knee joint in all types of ambulatory activities. The results of this study prove that this control method can predict the performance of a prosthesis and produce the required moment. Sup et al. (Sup et al., \u003cspan class=\"CitationRef\"\u003e2008b\u003c/span\u003e; Sup et al., \u003cspan class=\"CitationRef\"\u003e2007a\u003c/span\u003e) used this approach to manipulate the required torque at each joint during a single stride, represented as a piecewise series of impedance functions. For example, the stance flexion behavior of the knee can be approximated as a linear spring, whereas the pre-swing knee torque is described as a nonlinear spring. Thus, one can coordinate the prosthesis motion and control-level walking gait. Varol et al. (Varol et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e), also used a finite-state impedance strategy to execute the joint impedance according to the gait phase. In each phase, the knee and ankle torque was described by a passive spring and damper with a fixed equilibrium point and by shifting the joint impedances between the gait phases. For instance, shifting from a late state during walking to swing flexion occurs with the detection of toe-off. This structure indicates that the prosthesis is guaranteed to be passive within each gait phase and delivers power to the user only by shifting the linear stiffness and equilibrium point of the virtual spring between the phases. Therefore, prosthetic users can control prostheses directly and naturally. The results from this study demonstrated the robustness of the control system approach, which can be implemented in an active prosthetic leg for the user to perform various ambulatory movements more naturally.\u003c/p\u003e\n\u003ch2\u003e3.7 High Level Controller (Intent Recognition)\u003c/h2\u003e\n\u003cp\u003eA suitable input signal and fast machine-learning algorithm are required to design a good and reliable intent recognition system. Researchers have used various techniques to address this issue (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Determining the proper state transition within the locomotion modes (e.g., between the stance and swing phases of walking) is easier than defining the state transition between the locomotion modes (e.g., from the stance phase of walking to the swing phase of stair ascent), which is the reason why a high-level controller was developed. The purpose of having a high-level controller is to predict the intention of the prosthetic user, and this intent recognition is achieved through a pattern recognizer that compares the state of the prosthesis to probabilistic models of activity. Misclassification of intent recognition can cause imbalances among users. Several intent recognition strategies have been developed by researchers to be used as intent recognizer systems and are presented below.\u003c/p\u003e\n\u003ch2\u003e3.8 Locomotion Mode Recognition Strategy\u003c/h2\u003e\n\u003cp\u003eA more robust and accurate intent recognition system can be developed by classifying specific points during a single gait cycle. Previous studies by Young et al. (Young et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) and Varol et al. (Varol et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e) introduced data classifier windows based on each gait phase. A model of the group data was trained based on the sets of selected input sensors. Thus, the data were extracted to a certain frame length and converted into appropriate sets of features to be extracted from each window. After the input was selected, a set of training data was used to develop the model for each locomotion mode: walking, sitting, or standing. The model was established in real-time to determine the exact activity mode at a given instant. This strategy was proven successful but required a low-pass filter to improve the current mode determination in real-time. In contrast, Khademi et al. (Khademi et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e) used a gait mode recognition approach to provide maximum performance with minimum complexity for transfemoral amputees to control their prostheses. The purpose of introducing the new framework was to design a comprehensive control system that utilized multi-objective optimization. This was to find the optimal feature subsets that can produce an intention recognition system that is both precise and accurate. The results reported a 97% classification accuracy in detecting the locomotion mode with six subjects. Therefore, this study indicated the reliability and compactness of advanced optimization methods for locomotion mode detection implemented in active prosthetic legs. Yusof et al. (Yusof et al., 2018) proposed user gait mode recognition based on seven phases of classification from the in-socket sensory system that allows one to mimic a normal healthy walking gait. The purpose of the in-socket sensory system was to derive the user\u0026rsquo;s intention to actively actuate the knee and thus reduce the user\u0026rsquo;s metabolic energy consumption. A control framework approach using an adaptive neuro-fuzzy inference system (ANFIS) algorithm was proposed to decode the arrays from the in-socket sensory system and characterize it into the seven phases of gait states from heel strike to toe-off. Subsequently, the particular gait phase recognition was mapped to the knee joint actuator torque and cadence control output. The proposed control framework was validated using in-socket sensor signals from 30 gait cycles of a transfemoral amputee. The results showed 80\u0026ndash;90% accuracy in terms of gait kinematics and kinetic parameters. The ANFIS system has been proven to detect all seven phases of gait based on the amputee\u0026rsquo;s in-socket sensory system. However, although this study presented the reliability of the ANFIS control strategy as a control algorithm framework for user movement and intention detection, the authors have yet to verify its implementation in real prosthetic devices.\u003c/p\u003e\n\u003ch2\u003e3.9 Myoelectric Control Strategy\u003c/h2\u003e\n\u003cp\u003eResearchers have investigated and implemented surface electromyography (EMG) for intention detection to control the knee joint of a transfemoral prosthetic leg. Aeyels et al. (Aeyels et al., \u003cspan class=\"CitationRef\"\u003e1995\u003c/span\u003e) developed a computer-based control system using EMG surface sensors and an electrically modulated brake for gait mode identification. The EMG sensors act as switches or moderators to provide commands to the actuator system for the required gait mode. The major advantage of using EMG-based control is the ability of the amputee to move the prosthesis according to their desire, instead of limiting the motion to a certain point of locomotion function (Varol and Goldfarb, \u003cspan class=\"CitationRef\"\u003e2007a\u003c/span\u003e; Young et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Huang et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e; Furuya et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e; Alzaydi et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e; Bai et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; Delis et al., \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e; Huang et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e). Kevin et al. (Ha et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e) used a volitional control strategy during non-weight-bearing activities that utilized an impedance framework as its algorithm scheme. The joint was programmed with a given stiffness and damping value, which reflected the nominal impedance properties of an intact joint. Surface EMG placed on the user\u0026rsquo;s hamstring and quadriceps muscles of the residual limb acted as the set-point angle of the joint impedance. Rather than using EMG sensors only to command the flexion and extension of the knee, this study used a combination of quadratic discriminant analysis and principal analysis to align the user\u0026rsquo;s intent to flex or extend the knee joint. However, this approach is limited to non-weight-bearing activities; thus, it is not suitable for other locomotion settings. Wu et al. (Wu et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e) proposed the use of an EMG-based control approach for amputee users to directly control the active prosthetic leg using their muscles by activating the neural signal between the EMG sensors and the muscle itself. This mechanism of \u0026lsquo;active-reactive\u0026rsquo; control strategies mimics the actuation mechanism of a human biological joint. In addition, this study investigated the compatibility between EMG sensors and the amputee user\u0026rsquo;s skin while performing various strategies. Surface EMG sensors proved to be a non-intrusive interface for acquiring a user\u0026rsquo;s muscle-activating signals. The findings from the study indicated that the controller could predict the knee and ankle joint positions with the user\u0026rsquo;s neural command, which was similar to the standard gait of healthy subjects.\u003c/p\u003e\n\u003cp\u003eBrantley et al. (Brantley et al., \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e) on the other hand, utilized EMG sensors to extract the signal from the amputated limb using a nonlinear extension of the Kalman filter to infer the users\u0026rsquo; intended gait pattern and thus predict the kinematic values of the knee and ankle joints. Thus, neural machine interfaces can provide a more natural and continuous transition of gait during locomotion activities. This study specifically used the anterior and posterior muscles as indicators to predict the ankle and knee positions, which resulted in high accuracy. Moreover, this study provides insight into the improvement of closed-loop control of lower-limb prostheses when compared with other intent-recognition strategies. Despite all the interventions has been made in the prosthetic technology field, there is yet to be a commercialized prosthesis available in the market that utilizes EMG sensors. According to Aeyels et al. (Aeyels et al., \u003cspan class=\"CitationRef\"\u003e1995\u003c/span\u003e), there are several difficulties in implementing these strategies into commercialized ones, including identifying and capturing the EMG sensor signals, processing them, and characterizing them according to the prosthetic device orientation. The effects of muscle fatigue could also cause additional problems in the control system of the prostheses.\u003c/p\u003e\n\u003ch2\u003e3.10 Echo Control Strategy\u003c/h2\u003e\n\u003cp\u003eThe echo control strategy uses a sound limb kinetic profile that is modified to control the prosthesis side. Sensors were instrumented on the intact limb of unilateral amputees to capture its knee angle profile and adapted for use on the powered prosthesis side one-half cycle later. This was referred to as \u0026ldquo;modified trajectory echo control\u0026rdquo; in the development of an electro-hydraulically powered knee prosthesis (Varol and Goldfarb, \u003cspan class=\"CitationRef\"\u003e2007b\u003c/span\u003e; Flowers, \u003cspan class=\"CitationRef\"\u003e1971\u003c/span\u003e). The prosthetic limb mimicked the kinematics of the sound limb through minimal system training. This method required users to train and control their gait at an even number of steps. Ossur (OSSUR, USA) developed a commercialized self-contained powered knee prosthesis, which featured a concept similar to that of echo control, in which instrumented sensors were embedded on the sound limb. One drawback of this technique is that the sound-side leg must be instrumented with sensors. This requires users to don and doff additional instrumentation, which can be challenging for them to do. To don and doff the prosthetic leg itself requires considerable work and energy; thus, this technique is not practically convenient for amputee users. Although this technique is restricted to unilateral amputee users, it also requires the user to keep track of an \u0026lsquo;odd\u0026rsquo; number of steps, whereby an echoed step is undesirable. This forces the amputee to react to the limb, instead of interacting with it.\u003c/p\u003e\n\u003ch2\u003e3.11 Volitional Control Strategy\u003c/h2\u003e\n\u003cp\u003eThis control framework uses both mechanical and biological sensors to operate and is referred to as the biological-mechanical fusion system. Varol et al. (Varol et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e) utilized a Gaussian mixture model as its control framework to characterize the probability of the user and the prosthesis being engaged in a given activity mode. EMG sensors were used as the signal transmitter between the user and control system to generate complex classifier models that were simple for various locomotion modes (sitting, walking, standing, slopes, and climbing stairs). In addition, the control algorithm also generates the equilibrium point of angular velocity and joint impedance, which consists of a combination of joint stiffness and damping. Therefore, the knee can move accordingly with the correct joint output stiffness and damping to the desired position, thus providing a smooth transition of the gait between the user, prosthesis, and environment. These Gaussian mixture model classifiers behave similarly to linear discriminant analysis and quadratic discriminant analysis classifiers. Similarly, Huang et al. (Huang et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e) developed an algorithm based on neuromuscular-mechanical fusion to continuously recognize a variety of locomotion modes by prosthesis users. In the study, they used a support vector machine to decode the interfaces between the EMG sensors and user, thus providing the necessary data to the control system. The support vector machine technique has also been reported as a reliable classification method that provides better classification performance than any other artificial neural network for intended purposes. The findings from both of these studies can be used as guidance to further develop a volitional control framework for intention detection purposes of transfemoral prostheses.\u003c/p\u003e\n\u003ch2\u003e3.12 Performance Assessment\u003c/h2\u003e\n\u003cp\u003eAnother research line regarding the development of an active prosthetic leg is the performance validation of the prostheses. This could be related to their mobility capability in producing smooth transitions and the reliability of the prostheses in terms of performance and ambulation tasks, and an assessment of every prosthesis developed is required. Various assessment methods can be used to validate the performance, one of which is the use of motion analysis lab facilities, such as Vicon Nexus (Vicon Motion Systems Ltd., UK). A gait laboratory system provides detailed measurements of several human-body motion variables. This analysis allows researchers to obtain technical conclusions regarding the assessment of the prosthetic leg itself. Moreover, a gait lab can capture various joint movements. The reconstruction of the gait cycle is also possible. For instance, in prosthetic leg assessments, all angular parameters of the joint must be measured along with the ground reaction forces. The results obtained might be compared to the normal human leg database to determine how far the prosthetic leg performance is from an actual one with very high accuracy. Several studies have reported the use of gait analysis to study the performance of prostheses. For example, Jasni et al. (Jasni et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; Jasni et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e) used gait lab analysis to study the efficacy of the sensory system in a prosthetic leg socket. The results of this study verified the reliability of the piezoelectric sensors as an alternative control input using the force profile of the muscles and the corresponding ground reaction force while performing the movements. In addition, El-Sayed et al. (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) employed a gait lab to evaluate the performance of the transducer sensor feedback signal in controlling the movement of the prosthesis. In other words, both of these studies verified the feasibility of the sensors to be used as the input source signal for the control system of potential active prostheses by providing accurate signals of the voltage arrays. The studies presented and discussed the gait analysis results obtained from a transfemoral amputee who used their in-socket sensory system.\u003c/p\u003e\n\u003cp\u003eAnother way of assessing prosthetic leg performance is through comparison with other similar technologies. In this sense, Hafner et al. (Hafner et al., \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e) and Johansson et al. (Johansson et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e) conducted a comparative study to assess gait patterns and variable damping between mechanical and microprocessor knee devices. In these studies, metabolic data were collected from eight unilateral amputees while walking with the selected knee device to analyze the kinetics and kinematics of the prosthetic devices. A comparison of the metabolic rate energy of the users was also performed in these studies. This means that the metabolic consumption when using different prosthetic knee joints was measured and the efficiency of the design was obtained. Simulation studies have also been used as one method to assess the behavior of prosthetic legs. Yusof et al. (Yusof et al., 2018) used this method to validate a control algorithm for an active prosthetic leg. The purpose of using physical simulation is to estimate the dynamic behavior of the system and simulate the movement of the prosthetic leg. To begin the development of an actuated knee joint, simulation is recommended as the first-stage process to analyze whether the design or control framework can produce the desired ambulatory parameters. El-Sayed et al. (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) used this method to model and control the linear actuated transfemoral knee joint in basic daily movements. The simulation results presented a realistic simulation of the actuated mechanism in terms of the knee parameters. Other studies have gathered subjects\u0026rsquo; opinions for setting conclusions rather than using gait analysis alone to validate the performance of the prosthetic device. The qualitative results may provide better depth with regard to the users of the prosthesis itself. On the other hand, studies reported by Andrysek et al. (Andrysek et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e) evaluated the conceptual basis of the stance-phase controlled pediatric prosthetic knee joint design through a clinically tested prototype and used a questionnaire to analyze the efficacy of the knee system. Kirker et al. (Kirker et al., \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e) also used this method to evaluate the performance of different types of intelligent prosthetic knee devices. All prosthetic knee devices were assessed using a questionnaire given to six patients. Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003esummarizes the performance assessment tools used for the development of an active prosthetic leg.\u003c/p\u003e\n\u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSummary of performance assessments tool for development of an active prosthetic leg\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePerformance assessments validation\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAssessment tools\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAdvantages\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLimitations\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGait analysis lab (Jasni et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; Jasni et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVicon Nexus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAll the kinetics and kinematics parameter can be tested\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLimited to indoor activity testing\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative study (Hafner et al., \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e; Johansson et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMetabolic data with various variables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEasy to trace the problems in the prosthetic device and troubleshoot it\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLimited to commercialized prosthetic devices in the market\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSimulation study (El-Sayed et al., \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Mohd Yusof et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Bhakta et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMATLAB software, OpenSim simulations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCan simulate the movement of the device before fabrication stage, durable and safe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLimited to software analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurvey analysis (Andrysek et al., \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e; Kirker et al., \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e; Saglam et al., \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQuestionnaire\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEasy to handle\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eResults might be less scientific and valid, time consuming\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2\u003e3.13 Classification Table\u003c/h2\u003e\n\u003cp\u003eVarious control strategies can be implemented in prosthetic devices to achieve intention detection ability. Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eprovides a summary of previous studies conducted by other researchers regarding the criteria that must be considered to produce an active prosthetic leg.\u003c/p\u003e\n\u003cp\u003eTable 3. Summary of previous studies regarding the criteria needed for an active prosthesis development\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAuthor\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eProsthetic knee joint\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensory system \u0026amp; signal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eActuator characteristics (Types of motor\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePower rating\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eMoment)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eControl strategies\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStage\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePerformance assessments tools\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAmbulation tasks\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eUser-intention detection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo. of subjects\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAmmar et al.\u003c/strong\u003e (Alzaydi et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eActive prosthetic knee fuzzy logic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWireless (EMG and accelerometers)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eDC brushed motor\u003c/p\u003e\n \u003cp\u003e7648 rpm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eArtificial neural network\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment/ simulation/ Platform test design\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePlatform test design\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking only\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003eNone\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBellman et al. (\u003c/strong\u003eBellmann et al., 2010)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMicroprocessor knee (C-leg, Hybrid knee, Rheo knee, Adaptive 2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMechanical sensors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic actuator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state machine variable damping\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eCommercial in the market\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait laboratory (Optoelectronic camera system combined with 2 force plates)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol et al.\u003c/strong\u003e (Varol et al., 2010)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLoad cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eElectric motor, lithium battery 118 W\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLocomotion mode recognition + finite state-based impedance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking, sitting, and standing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYoung et al.\u003c/strong\u003e (Young et al., 2014)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered prosthesis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eAccelerometers + gyroscopes + load cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eRobotic leg built by Vanderbilt University\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state machine control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking, sit to stand, ramp and slope\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol et al.\u003c/strong\u003e (Varol et al., 2009)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLoad cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eElectric motor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait mode intent recognition\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis + simulation analysis for control system purposes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking at different speeds, standing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJohansson et al.\u003c/strong\u003e (Johansson et al., 2005)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eC-leg, Rheo knee, Mauch knee\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEMG + accelerometers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic motor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment+ already commercialize in the market\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGround level of walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDawley et al.\u003c/strong\u003e (Dawley et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee with impedance control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEMG+ uniaxial load cell\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eDual regenerative servo amplifiers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance-based control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalk in multi-terrain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHuang et al.\u003c/strong\u003e (Huang et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic passive knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEMG + load cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic servo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePattern classification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking, stair ascent and descent, obstacles, ramp\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYoung et al.\u003c/strong\u003e (Young et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e16 mechanical sensors (load cells, potentiometers, encoders)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eDouble acting pneumatic actuator, four-way servo valves,\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance based control using time history information\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGround level walking, ramp walking 10-degree slope, ascend descend stairs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.945802337938364%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSup et al.\u003c/strong\u003e (Sup et al., 2007b)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eLoad cells, uniaxial load cells, joint motion sensors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eDouble acting pneumatic actuator, four-way servo valves, lithium battery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance based control approach\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eWalking at a different speed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.883103081827842%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.989373007438894%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJasni et al.\u003c/strong\u003e (Jasni et al., 2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePiezoelectric sensors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic actuator (OTTOBOCK)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance based control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalking different speed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEl-Sayed et al.\u003c/strong\u003e (El-Sayed et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eVariable impedance control approach\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSimulation software\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSimulation software\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalking at a different speed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003eNone\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWu et al.\u003c/strong\u003e (Wu et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eActive above-knee prostheses\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eEMG + joint angle (uniaxial load cell + rotary potentiometer)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eDC motor+ ball screw assembly\u003c/p\u003e\n \u003cp\u003e150 W\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eEcho + impedance\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiments/ lab testing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSet of free swing experiments + set of level walking experiments\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHa et al.\u003c/strong\u003e (Ha et al., 2011)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eEMG, uniaxial load cells (ELPF-500L), potentiometer.\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(ALPS RDC503013), moment sensor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eMaxon EC30 Powermax brushless motor, 200 W, lithium polymer battery, 29.6 V, nominal rating 4000 mA.h capacity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance-based weight-bearing control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab simulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eMATLAB real-time workshop\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSit to stand only\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eKhademi et al.\u003c/strong\u003e (Khademi et al., 2019)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eOTTOBOCK prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSensors from the Vicon system camera\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eOTTOBOCK actuator system\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait mode recognition control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment + simulation analysis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis using 16-camera Vicon Nexus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGround walking, standing, and walking at various speeds.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol and Goldfarb\u003c/strong\u003e (Varol and Goldfarb, 2007a)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLoad cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eServo-valve\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLocomotion mode recognition + finite state-based impedance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalking, sitting, and standing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBrantley et al.\u003c/strong\u003e (Brantley et al., 2017)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eNot stated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eNot stated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state control approach\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalk on a multi-terrain gait course (stairs, ramp, and ground-level)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDelis et al.\u003c/strong\u003e (Delis et al., 2009)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eMicro-controlled prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eEMG, accelerometers, Kalman filter.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003ePattern recognition\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis \u0026amp; MATLAB simulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalking\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e4 able-bodied individuals\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.040339702760084%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrimmer et al.\u003c/strong\u003e (Grimmer et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eActive prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eForce sensors, infrared cameras\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eSeries elastic actuator (SEA) with spring\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state impedance control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eWalking and running\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.872611464968152%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978768577494693%\" valign=\"top\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYusof et al.\u003c/strong\u003e (Yusof et al., 2018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePiezoelectric sensors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state impedance control (ANFIS)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking at different speeds\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBai et al.\u003c/strong\u003e (Bai et al., 2016\u0026ndash;)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEEG and switch\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eSingle axis knee and manual locking system\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eVolitional control of a knee-locker switch in real time\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eShort distance of ground level walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eThatte and Geyer\u0026nbsp;\u003c/strong\u003e(Thatte and Geyer, 2016)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eActive knee SEA unit\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eIMU and load cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHybrid neuromuscular model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eSimulation experiments\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHargrove et al\u003c/strong\u003e. (Hargrove et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEMG\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee by Vanderbilt- University\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eVolitional impedance control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eSitting only\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEl-Sayed et al.\u003c/strong\u003e (El-Sayed et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMechanical prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePiezoelectric\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMechanical knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePfeifer et al.\u003c/strong\u003e (Pfeifer et al., 2015)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eTethered powered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLoad cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eVisco elastic actuator (SVA), Maxon EC30 4-pole\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eFinite state control strategy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePilot experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking at different speeds\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHargrove et al.\u003c/strong\u003e (Hargrove et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered prosthesis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eEMG\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMaxon EC30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePattern recognition strategy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eVirtual environment experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eSitting\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJasni et al.\u003c/strong\u003e (Jasni et al., 2019)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePiezoelectric sensors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eHydraulic actuator (OTTOBOCK)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eImpedance based control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eWalking different speed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWu et al.\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Wu et al., 2021)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ePotentiometer+ EMG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003ea slider-crank mechanism with electric DC motor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eReinforcement learning algorithm.\u0026nbsp;Finite state machine\u0026nbsp;+ impedance control +\u0026nbsp;intact\u0026nbsp;knee motion tracking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis \u0026amp; OpenSim simulations\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGround-level-walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e2 able- bodies\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eShaikh \u0026amp; Malhotra\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Shaikh \u0026amp; Malhotra, 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMicro-controlled prosthesis\u003c/p\u003e\n \u003cp\u003e(Microcontroller LPC2148)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eSensory feedback\u0026nbsp;\u003c/p\u003e\n \u003cp\u003esystem IMU + 3-axis gyroscope\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003elinear actuator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eMotion Fusion Algorithm\u0026nbsp;\u003c/p\u003e\n \u003cp\u003ewith real-time feedback.\u003c/p\u003e\n \u003cp\u003echip (PSoC 4), 042-BLE, with Bluetooth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eDifferent speed of walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.978540772532188%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.085836909871245%\" valign=\"top\"\u003e\n \u003cp\u003e6 able-bodied individuals\u003c/p\u003e\n \u003cp\u003e14 TF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWen et al.\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Wen et al., 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003ePersonalized robotic knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eload cell+ angle sensor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\"\u003e\n \u003cp\u003eDC motor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eApproximate Dynamic Programming (ADP)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e+Finite-state machine impedance control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis \u0026amp; OpenSim simulations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.884368308351178%\" valign=\"top\"\u003e\n \u003cp\u003eGround-level-walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.635974304068522%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e1 able-bodied individual\u003c/p\u003e\n \u003cp\u003e1 TF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e\u003cstrong\u003eKadhim et al.\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Kadhim et al., 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eaxis force sensors+\u0026nbsp;pressure sensors platform\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\"\u003e\n \u003cp\u003eelectric DC motor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eAdaptive Neuro-Fuzzy Inference System (ANFIS)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.884368308351178%\" valign=\"top\"\u003e\n \u003cp\u003eGround-level-walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.635974304068522%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e\u003cstrong\u003eLiu et al.\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Liu et al., 2017)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003epowered prosthetic knee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eEMG (MA300-XVI)+\u0026nbsp;load cell (Mini58, ATI, Apex, NC, USA)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\"\u003e\n \u003cp\u003ePAMs linear actuator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003epattern recognition (PR) + LMR\u0026nbsp;(locomotion mode recognition) +\u0026nbsp;finite-state impedance controller\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.884368308351178%\" valign=\"top\"\u003e\n \u003cp\u003eLevel walking and ascending/descending slopes.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.635974304068522%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e2 able-bodied individuals\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2TF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e\u003cstrong\u003eBhakta et al.\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e(Bhakta et al., 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003ePowered knee and ankle prosthesis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eload cell+ embedded sensor encoders\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\"\u003e\n \u003cp\u003e2 brushless DC motors +\u0026nbsp;gear transmission+\u0026nbsp;harmonic drive+\u0026nbsp;Li-Po battery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eimpedance control + Finite-state machine impedance control+ Pattern recognition strategy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eLab experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.850107066381156%\" valign=\"top\"\u003e\n \u003cp\u003eGait lab analysis \u0026amp; MATLAB data analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.884368308351178%\" valign=\"top\"\u003e\n \u003cp\u003eLevel walking and ascending/descending slopes.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.635974304068522%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.74304068522484%\" valign=\"top\" \u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable 4. Methodology quality of included studies according to downs and black checklist\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy Criteria\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e17\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e20\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e27\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eQuality score\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePercentage (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBellmann et al.\u0026nbsp;\u003c/strong\u003e(Bellmann et al., 2010)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10/14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e71.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAmmar et al.\u0026nbsp;\u003c/strong\u003e(Alzaydi et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDelis et al.\u0026nbsp;\u003c/strong\u003e(Delis et al., 2009)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e16/17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrimmer et al.\u003c/strong\u003e (Grimmer et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12/13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e92.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePfeifer et al.\u0026nbsp;\u003c/strong\u003e(Pfeifer et al., 2015)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e15/16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e93.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHargrove et al.\u0026nbsp;\u003c/strong\u003e(Hargrove et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e93.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYusof et al.\u003c/strong\u003e (Yusof et al., 2018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e16/17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol et al.\u003c/strong\u003e (Varol et al., 2010)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e89.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYoung et al.\u0026nbsp;\u003c/strong\u003e(Young et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e18/22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e81.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYoung \u0026amp; Hargrove\u003c/strong\u003e (Young \u0026amp; Hargrove, 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e18/21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e85.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eThatte \u0026amp; Geyer\u0026nbsp;\u003c/strong\u003e(Thatte \u0026amp; Geyer, 2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9/12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e75.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDawley et al.\u0026nbsp;\u003c/strong\u003e(Dawley et al., 2013)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5/8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e62.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eYoung et al.\u0026nbsp;\u003c/strong\u003e(Young et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10/14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e71.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol et al.\u0026nbsp;\u003c/strong\u003e(Varol et al., 2009)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e15/16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e93.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFite et al.\u0026nbsp;\u003c/strong\u003e(Fite et al., 2007)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8/17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e47.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBrantley et al.\u0026nbsp;\u003c/strong\u003e(Brantley et al., 2017)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e89.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJasni et al.\u003c/strong\u003e (Jasni et al., 2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e89.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eKevin et al.\u0026nbsp;\u003c/strong\u003e(Kevin et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e15/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e100.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDelis et al.\u0026nbsp;\u003c/strong\u003e(Delis et al., 2009)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14/16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e87.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVarol \u0026amp; Goldfarb\u0026nbsp;\u003c/strong\u003e(Varol \u0026amp; Goldfarb, 2007a)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10/11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e90.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBai et al. 2015\u003c/strong\u003e (Bai et al., 2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14/16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e87.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJohansson et al.\u0026nbsp;\u003c/strong\u003e(Johansson et al., 2005)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e40.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEl-Sayed et al.\u0026nbsp;\u003c/strong\u003e(El-Sayed et al., 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e16/18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e88.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWu et al.\u0026nbsp;\u003c/strong\u003e(Wu et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6/10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e60.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSup et al.\u003c/strong\u003e (Sup et al., 2007b)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGholamreza et al.\u003c/strong\u003e (Khademi et al., 2019)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e93.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHuang et al.\u003c/strong\u003e (Huang et al., 2011)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10/14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e71.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJasni et al.\u0026nbsp;\u003c/strong\u003e(Jasni et al., 2019)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e89.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eJasni et al.\u0026nbsp;\u003c/strong\u003e(Jasni et al., 2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17/18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e94.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBhakta et al.\u0026nbsp;\u003c/strong\u003e(Bhakta et al., 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eLiu et al.\u0026nbsp;\u003c/strong\u003e(Liu et al., 2017)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWu et al.\u0026nbsp;\u003c/strong\u003e(Wu et al., 2021)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eShaikh\u003c/strong\u003e \u003cstrong\u003e\u0026amp; Malhotra\u003c/strong\u003e (Shaikh \u0026amp; Malhotra, 2020)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eWen et al.\u003c/strong\u003e (Wen et al., 2020)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eKadhim et al.\u0026nbsp;\u003c/strong\u003e(Kadhim et al., 2020)\u003cstrong\u003e\u0026nbsp;\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13/15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e* Items related to external validity (items 11, 12, 13), internal validity (items 14, 15, 16, 17), and internal validity \u0026ndash; confounding selection bias (items 21, 22, 23, 24, and 26) from the Downs and Black (D\u0026amp;B) criteria (D. a Winter, and a O, 1975) were not applicable in most studies for the present review, being applied only to evaluate randomized controlled trial reports. Abbreviations: D\u0026amp;B criteria met = 1, D\u0026amp;B criteria unmet = 0, Unable to determine = 0, Criteria not applicable to the study = NA. \u0026ldquo;Downs and Black\u0026rdquo; \u0026ndash; Criteria are as summarized follows: 1-Hypothesis stated, 2- Outcome described in Introduction/ Method, 3- Participants\u0026rsquo; characteristics described, 4- Intervention described, 5- Principal confounder in each group subjects described, 6- Findings described, 7- Data distribution reporting, 8- Description of adverse events, 9- Characteristics of patients lost to follow up described, 10- Exact p-values reported, 18- Statistical tests used to assess the main outcomes reported, 19- Adherence to the intervention described, 20- Accuracy of outcome measures described, 25- Adjustment for confounding in the analyses from which the main findings are reported.\u003c/p\u003e"},{"header":"4 Conclusion","content":"\u003cp\u003eThe control interface mechanisms for intention detection that may be used in an active prosthetic limb are realistically scientifically supported in this work. The quality of the studies varied from acceptable to good, with scores between 40% and 100% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e4\u003c/span\u003e), with a mean score of 83.3%, which is regarded as excellent. This is according to the accepted D\u0026amp;B checklist (AT, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). All studies accurately described the research's purpose and underlying premise (criteria 1) as well as the reliability of its outcome measures (criteria 20). Overall, the results of these investigations came from trustworthy lab tests. The paucity of application of the probability value results indicates that few of them provided clinically pertinent data. Most D\u0026amp;B criterion items demonstrate how few finds and advances are generally practical. The user's-biological-input focused sensory system in microprocessor active prosthetic legs is the best way to identify an amputee user's gait intention, it can be confidently inferred based on the quality outcome of this review. This study underlines and draws attention to many key and significant lines of research that have been presented throughout the years:\u003c/p\u003e \u003cp\u003e \u003col style=\"list-style-type:upper-roman;\"\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eResearch lines examining several intention-detection sensors that can be used into active transfemoral prosthetic legs\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eResearch focuses on the different feedback and control techniques utilized for intention detection.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eResearch areas focused on evaluating the effectiveness of active transfemoral prosthetic legs.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe first study line explored the numerous sensory systems that, depending on the requirements needed by the researcher to construct the prosthetic leg, might be included into the active prosthetic leg for intention detection reasons. A prosthetic limb device's control architecture framework methodologies were covered in the second study area. Depending on the prosthesis functionality and the kinds of sensory systems used in the prosthetic legs, several control interface frameworks are established for intention detection. Given the available actuators for active prosthesis, weight-vs-size incompatibility has proven to be a design constraint. This highlights the necessity of creating lighter parts, such as titanium as its main material for prosthetic knee joint design or other comparable possibilities, or concentrating on the creation of artificial muscles. In order to limit muscular activity and make up for the energy users waste when using the prosthetic device, the usage of EMG sensors was highlighted. Nevertheless, despite the present improvements in prosthetic technology, further research is still required to develop more effective processing methods for recognizing and filtering biological signals. The last claim relates to the performance evaluation of a certain prosthesis to confirm the caliber of prosthetic devices. The majority of research employed motion lab analysis to collect certain metrics in real time, which were used as conclusion-making judgement criteria. Only commercially accessible technologies have been utilized for actual environmental evaluations. The ability to operate fitted prosthetic devices would be considerably improved if all three issues were addressed during technological development.\u003c/p\u003e \u003cp\u003eThe type of sensory system utilized in prosthetic legs affects the control algorithms employed in such legs. For instance, the mid-level controller often conducts the identification of the present state that the device is in while controlling the prosthetic device-oriented sensory system. Data from the prosthetic device itself will be collected by the sensors, which will then send it to the controller for interpretation before sending the algorithm to the actuator system. While this was going on, the user-biological-oriented sensory system and the neuro-mechanical fusion sensory system were both controlled by two levels of controllers. The high-level controller (fuzzy) would collect the signal from the sensory system (EMG or EEG) and analyze it before sending it to the low-level controller (PID) to instruct the actuator of the prosthetic leg system. Additionally, several methodologies might be employed to verify the effectiveness of each sensory system used in the prosthetic legs. One of these methods involves utilizing a motion analysis device to record every kinematic and kinetic parameter as amputees walked on prosthetic legs. To confirm the dependability of the sensors, the output signals from the sensory system will be mapped with the output from the motion analysis system. To assess the functioning of an active prosthetic leg's sensory system, the majority of research typically utilized five to 10 individuals. The tendency of correctness of the research study's sensory system increases with the number of people engaged.\u003c/p\u003e \u003cp\u003eFurther examination of the research shown in Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e3\u003c/span\u003e reveals that, according to the above-mentioned needed criteria, investigations involving the prosthetic-device oriented sensory system were more developed than those involving the other two options. A few investigations, including those on the CYBERLEGS and VI Knee reported using the produced active transfemoral prosthetic leg prototype, which was outfitted with and operated using the suggested sensory system. It demonstrates the researchers' faith in the suggested system's ability to keep the individuals safe. The user's-biological-input-oriented and neuro-mechanical fusion sensory system research is still in its early stages.\u003c/p\u003e \u003cp\u003eIt has been demonstrated that the impedance-based control method effectively coordinates the movements of the prosthetic device-oriented sensory system. Additionally, it was shown to be employed as an input signal to distinguish between different ambulation modes. However, the evaluation reveals that only one study reported using a kinematic sensor exclusively to carry out the job, whereas the others reported using both kinematic and kinetic sensors to detect activity mode transitions. However, as of now, this kind of sensory system's effectiveness is still insufficient to be employed in determining the user's purpose. This is due to the fact that there is no direct contact between the sensors and the user's body, and it is also unable to access the user's neurological input, which has been shown to be capable of deciphering motion intent before a movement really occurs.\u003c/p\u003e \u003cp\u003eAdditionally, the sensory system that is focused on prosthetic devices does not incorporate information about environmental responses. For example, an able-bodied person's movement pattern is also influenced by information from the sensory organs (such as the eyes and hearing) that can detect obstacles in the way. In order to avoid the obstruction, the brain will tell the muscles to flex their knees more and swing for a longer amount of time. The safety of the prosthesis user might be jeopardized in the absence of this information. In their study, Martin et al. asserted that if there is a discrepancy between the user's planned motion and the motion established by the prosthetic device controller, it may endanger the user's safety, cause gait deficit, and ultimately result in prosthetic device rejection. Using a sensory system that gathers data from the user's bodily input, such as EEG and EMG, is one technique to prevent this.\u003c/p\u003e \u003cp\u003eEMG electrodes are used more often in lower-limb prosthetic devices. Even though several research showed that the EMG signal is reliable in determining the user's intention prior to the action actually occurring, its performance is constrained by a few limitations. It is extremely vulnerable to electrode-skin conductivity, motion artefacts, electrode misalignment, and intermuscular communication. Additionally, the EMG sensing system's don and doff procedures are laborious. Therefore, it can be projected that an EMG-only based sensory system will not be compelling enough to be employed in lower-limb MPC prosthetic until these signal-quality related concerns are properly solved.\u003c/p\u003e \u003cp\u003eThe controller receives more detailed information from the fusion sensory system. This sort of sensory system can measure muscle activation as well as kinetic and kinematic data, which enables the development of robust control algorithms and smooth switching between ambulation modes for MPC prostheses. In terms of processing time and RAM and ROM capacity of the processor, the complexity of the system to handle heavy information from the sensors (such as EMG and mechanical sensors) may limit the usefulness of the sensory system. According to research by Zhang et al., kinematics information was not as helpful as EMG and GRF/moment data for TF amputees doing real-time intent detection tasks. However, a more recent work by Stolyarov et al. shown that the intent recognition task could be accomplished using simply inertial sensors and the translational motion tracking approach.\u003c/p\u003e \u003cp\u003eThe sensory and a control system of a microprocessor-controlled prosthetic limb were reviewed in this literature study's conclusion. The classification of different sensory and control systems into new categories was put forth. In essence, depending on what researchers require in the prosthetic limb system, each form of sensor and its control has advantages and disadvantages. The sensory system designed for prosthetics may operate more effectively and would be useful for controlling how the device moves. Additionally, researchers working on prosthetic devices frequently employ this kind of sensory system, which is currently available on the market. This sensory system's shortcomings include a decreased ability to foretell users' intentions. Studies that employ user-biologically oriented input are effective in identifying users' intentions, but they lack dependability in real-world settings. One of the identified issues with using these kinds of sensors is that the output signal's resilience isn't compelling enough to be employed in active transfemoral prostheses. Fusion sensory systems are the finest choice of sensors for upcoming studies since they are reliable in giving the controller adequate data because they employ both mechanical and user-biological sensors. Additionally, when used in conjunction with an effective control system, these sensors might offer a smooth transition mode for active transfemoral prostheses. Therefore, our investigation revealed that the adoption of a simpler sensory system to operate an active transfemoral prosthetic limb is feasible with an inventive strategy to extrapolate the limited sensor information into a broader group of useful information. As a result, the two requirements for an effective sensory system\u0026mdash;practicality and quality of the sensory data\u0026mdash;can be met. Given the linked biomechanics of human gait, which the brain regulates subconsciously, it is regarded as a complicated activity. The difficulty in prosthetic technology still exists, which explains why no prosthetic device has been able to replace a human limb in the same way that the original one would function. Even with something as straightforward as walking, the difficulty comprises several elements that may be divided into various study areas. No matter how modest, this requires contributions from many fields of expertise.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eD\u0026amp;B\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Down and Black Checklist\u003c/p\u003e\n\u003cp\u003eEMG\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Electromyography\u003c/p\u003e\n\u003cp\u003eCC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Classical Control\u003c/p\u003e\n\u003cp\u003eANFIS\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Adaptive Neuro-fuzzy Inference System\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgment\u003c/p\u003e\n\u003cp\u003eThis research was partially supported by the University Malaya Faculty Research Grant (GPF068A-2018). The first author received a scholarship from the Department of Public Services, Malaysia (JPA Scholarship).\u003c/p\u003e\n\u003cp\u003eConflict of Interest\u003c/p\u003e\n\u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u003c/p\u003e\n\u003cp\u003eAuthor Contributions\u003c/p\u003e\n\u003cp\u003eNur Hidayah: Conceptualization, Methodology, Data curation, Data analysis, Writing \u0026ndash; Original draft preparation. Nur Azah.: Data curation, Writing- review \u0026amp; editing, Supervision, Funding acquisition. Farahiyah: Data curation. Khin Wee: Data curation, Fanny Oddon \u0026ndash; literature search and manuscript writing.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eGeran Penyelidikan Fakulti - Faculty Research Grant code GPF068A-2018 and Jabatan Pertahanan Awam (JPA) Scholarships.\u003c/p\u003e\n\u003cp\u003eData Availability Statement\u003c/p\u003e\n\u003cp\u003eNo datasets were generated or analyzed during the current study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAeyels, B., Van Petegem, W., Sloten, J. V., Van Der Perre, G., Peeraer, L. (1995). An EMG-based finite state approach for a microcomputer-controlled above-knee prosthesis. Proc. 17th Int. Conf. Eng. Med. Biol. Soc. 2, 1315\u0026ndash;1316.\u003c/li\u003e\n\u003cli\u003eAlzaydi, A. A., Cheung, A., Joshi, N., Wong, S. (2011). Active prosthetic knee fuzzy logic-PID motion control, sensors and test platform design. Int. J. Sci. Eng. Res. 2, 1\u0026ndash;17.\u003c/li\u003e\n\u003cli\u003eAndrysek, J., Naumann, S., Cleghorn, W. L. (2005). Design and quantitative evaluation of a stance-phase controlled prosthetic knee joint for children. IEEE Trans. Neural Syst. Rehabil. Eng. 13, 437\u0026ndash;443. doi: 10.1109/TNSRE.2005.856071.\u003c/li\u003e\n\u003cli\u003eAT. (2008). Studies. Downs Black Checklist.\u003c/li\u003e\n\u003cli\u003eAu, S., Berniker, M., Herr, H. (2008). Powered ankle-foot prosthesis to assist level-ground and stair-descent gaits. Neural Netw. 21, 654\u0026ndash;666. doi: 10.1016/j.neunet.2008.03.006.\u003c/li\u003e\n\u003cli\u003eBabu\u0026scaron;ka, R., Verbruggen, H. B. (1996). An overview of fuzzy modeling for control. Control Eng. Pract. 4, 1593\u0026ndash;1606. doi: 10.1016/0967-0661(96)00175-X.\u003c/li\u003e\n\u003cli\u003eBai, O., Kelly, Graham., Ding, Yu Fei., Douglas, Murphy., John, Fox., Brian, Burkhardt., et al. (2016). A wireless, smart EEG system for volitional control of lower-limb prosthesis. IEEE Reg. 10 Annu. Int. Conf. Proc. TENCON, 2016, 1\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eBellmann, M., Schmalz, T., Blumentritt, S. (2010). Comparative biomechanical analysis of current microprocessor-controlled prosthetic knee joints. Arch. Phys. Med. Rehabil. 91, 644\u0026ndash;652. doi: 10.1016/j.apmr.2009.12.014.\u003c/li\u003e\n\u003cli\u003eBhakta, K., Camargo, J., Kunapuli, P., Childers, L., Young, A. (2020). Impedance control strategies for enhancing sloped and level walking capabilities for individuals with transfemoral amputation using a powered multi-joint prosthesis. Mil. Med. 185, 490\u0026ndash;499. doi: 10.1093/milmed/usz229.\u003c/li\u003e\n\u003cli\u003eBorjian, R., Lim, J., Khamesee, M. B., Melek, W. (2008). The design of an intelligent mechanical active prosthetic knee. IECON 34th Annu. Conf. Ieee Ind. Electron. Soc. Vols 1, 5, Proc. 2918\u0026ndash;2923.\u003c/li\u003e\n\u003cli\u003eBrantley, J. A., Luu, T. P., Nakagome, S., Contreras-Vidal, J. L. (2017). Prediction of lower-limb joint kinematics from surface EMG during overground locomotion IEEE Int. Conf. Syst. Man, Cybern. SMC 2017, 1705\u0026ndash;1709.\u003c/li\u003e\n\u003cli\u003eCelebi, B., Yalcin, M., Patoglu, V. (2013). AssistOn-Knee: A self-aligning knee exoskeleton, in IEEE Int. Conf. Intell. Robots Syst. 996\u0026ndash;1002.\u003c/li\u003e\n\u003cli\u003eDawley, J. A., Fite, K. B., Fulk, G. D. (2013). EMG control of a bionic knee prosthesis: Exploiting muscle co-contractions for improved locomotor function. IEEE Int. Conf. Rehabil. Robot.. 2013, 6650389. doi: 10.1109/ICORR.2013.6650389.\u003c/li\u003e\n\u003cli\u003eDelis, A. L., Carvalho, J. L. A., Da Rocha, A. F., Ferreira, R. U., Rodrigues, S. S., Borges, G. A. (2009). Estimation of the knee joint angle from surface electromyographic signals for active control of leg prostheses. Physiol. Meas. 30, 931\u0026ndash;946. doi: 10.1088/0967-3334/30/9/005.\u003c/li\u003e\n\u003cli\u003eDowns, S. H., Black, N. (1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J. Epidemiol. Community Health. 52, 377\u0026ndash;384. doi: 10.1136/jech.52.6.377.\u003c/li\u003e\n\u003cli\u003eDyck, W. R., Onyshko, S., Hobson, D. A., Winter, D. A., Quanbury, A. O.W. R. Dyck, S. Onyshko, D., D. a Winter, and a O. (1975). Bull. Prosthet. Res. 169\u0026ndash;186.\u003c/li\u003e\n\u003cli\u003eEl-Sayed, A. M., Hamzaid, N. A., Abu Osman, N. A. A. (2014). Piezoelectric bimorphs \u0026ldquo;characteristics as in-socket sensors for transfemoral amputees\u0026rdquo;. Sensors. 14, 23724\u0026ndash;23741. doi: 10.3390/s141223724.\u003c/li\u003e\n\u003cli\u003eEl-Sayed, A. M., Hamzaid, N. A., Azuan, N., Osman, A. (2014). Applied mathematics\u0026amp; information sciences modelling and control of a linear actuated transfemoral knee joint in basic daily movements. Appl. Math. Inf. Sci. 7, 1\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003eFite, K., Mitchell, J., Sup, F., Goldfarb, M. (2007). Design and control of an electrically powered knee prosthesis, in 10th International Conference on Rehabilitation Robotics, ICORR\u0026rsquo;07 (IEEE Publications).\u003c/li\u003e\n\u003cli\u003eFlowers, W. C. (1971). \u0026lsquo;A man-interactive simulator system for an above-knee prosthetics studies,\u0026rsquo; Massachusettes Institute of Technology.\u003c/li\u003e\n\u003cli\u003eFuruya, Y., Mikami, T., Suzuki, T., Fujimoto, Y. (2013). On an active prosthetic knee joint driven by a high thrust force helical motor. IECON Proc. (Ind. Electron. Conf.). 2, 5894\u0026ndash;5899.\u003c/li\u003e\n\u003cli\u003eGrimmer, M., Eslamy, M., Seyfarth, A. (2014). Energetic and peak power advantages of series elastic actuators in an actuated prosthetic leg for walking and running. Actuators. 3, 1\u0026ndash;19. doi: 10.3390/act3010001.\u003c/li\u003e\n\u003cli\u003eHa, K. H., Varol, H. A., Goldfarb, M. (2011). Volitional control of a prosthetic knee using surface electromyography. IEEE Trans. Biomed. Eng. 58, 144\u0026ndash;151. doi: 10.1109/TBME.2010.2070840.\u003c/li\u003e\n\u003cli\u003eHafner, B. J., Askew, R. L. (2015). Physical performance and self-report outcomes associated with use of passive, adaptive, and active prosthetic knees in persons with unilateral, transfemoral amputation: Randomized crossover trial. J. Rehabil. Res. Dev. 52, 677\u0026ndash;700. doi: 10.1682/JRRD.2014.09.0210.\u003c/li\u003e\n\u003cli\u003eHafner, B. J., Willingham, L. L., Buell, N. C., Allyn, K. J., Smith, D. G. (2007). Evaluation of function, performance, and preference as transfemoral amputees transition from mechanical to microprocessor control of the prosthetic knee. Arch. Phys. Med. Rehabil. 88, 207\u0026ndash;217. doi: 10.1016/j.apmr.2006.10.030.\u003c/li\u003e\n\u003cli\u003eHamzaid, N. A., Mohd Yusof, N. H., Jasni, F. (2020). Sensory systems in micro-processor controlled prosthetic leg: A review. IEEE Sensors J. 20, 4544\u0026ndash;4554. doi: 10.1109/JSEN.2019.2944653.\u003c/li\u003e\n\u003cli\u003eHargrove, L. J., Simon, A. M., Lipschutz, R., Finucane, S. B., Kuiken, T. A. (2013). Non-weight-bearing neural control of a powered transfemoral prosthesis. J. Neuroeng. Rehabil. 10, 62. doi: 10.1186/1743-0003-10-62.\u003c/li\u003e\n\u003cli\u003eHargrove, L. J., Simon, A. M., Young, A. J., Lipschutz, R. D., Finucane, S. B., Smith, D. G. et al. (2013). Robotic leg control with EMG decoding in an amputee with nerve transfers. N. Engl. J. Med. 369, 1237\u0026ndash;1242. doi: 10.1056/NEJMoa1300126.\u003c/li\u003e\n\u003cli\u003eHuang, H., Zhang, F., Hargrove, L. J., Dou, Z., Rogers, D. R., Englehart, K. B. (2011). Continuous locomotion-mode identification for prosthetic legs based on neuromuscular\u0026ndash;mechanical fusion. IEEE Trans. Biomed. Eng. 58, 2867\u0026ndash;2875. doi: 10.1109/TBME.2011.2161671.\u003c/li\u003e\n\u003cli\u003eHuang, H., Zhang, F., Sun, Y. L., He, H. (2010). Design of a robust EMG sensing interface for pattern classification. J. Neural Eng. 7, 056005. doi: 10.1088/1741-2560/7/5/056005.\u003c/li\u003e\n\u003cli\u003eJasni, F., Hamzaid, N. A., Al-Nusairi, T. Y., Mohd Yusof, N. H. Mohd., Shasmin, H. N., Ng, S. C. (2019). Feasibility of a gait phase identification tool for transfemoral amputees using piezoelectric- based in-socket sensory system. IEEE Sensors J. 19, 6437\u0026ndash;6444. doi: 10.1109/JSEN.2019.2910161.\u003c/li\u003e\n\u003cli\u003eJasni, F., Hamzaid, N. A., Muthalif, A. G. A., Zakaria, Z., Shasmin, H. N., Ng, S. (2016). In-socket sensory system for transfemoral amputees using piezoelectric sensors: An efficacy study. IEEE/ASME Trans. Mechatron. 21, 2466\u0026ndash;2476. doi: 10.1109/TMECH.2016.2578679.\u003c/li\u003e\n\u003cli\u003eJasni, F., Hamzaid, N. A., T. Y. Al-nusairi, N. Hidayah, and M. Yusof. Feasibility of a gait phase identification tool for transfemoral amputees using piezoelectric- based in-socket sensory system. (2019).\u003c/li\u003e\n\u003cli\u003eJohansson, J. L., Sherrill, D. M., Riley, P. O., Bonato, P., Herr, H. (2005). A clinical comparison of variable-damping and mechanically passive prosthetic knee devices. Am. J. Phys. Med. Rehabil. 84, 563\u0026ndash;575. doi: 10.1097/01.phm.0000174665.74933.0b.\u003c/li\u003e\n\u003cli\u003eKadhim, D. A., Raheema, M. N., Hussein, J. S. (2020). Design of an Intelligent Controller for above Knee Prostheses based on an Adaptive Neuro-Fuzzy Inference System. IOP Conf. Ser.: Mater. Sci. Eng. 671. doi: 10.1088/1757-899X/671/1/012066.\u003c/li\u003e\n\u003cli\u003eKhademi, G., Mohammadi, H., Simon, D. (2019). Gradient-based multi-objective feature selection for gait mode recognition of transfemoral amputees. Sensors (Basel). 19. doi: 10.3390/s19020253.\u003c/li\u003e\n\u003cli\u003eKirker, S., Keymer, S., Talbot, J., Lachmann, S. (1996). An assessment of the intelligent knee prosthesis. Clin. Rehabil. 10, 267\u0026ndash;273. doi: 10.1177/026921559601000314.\u003c/li\u003e\n\u003cli\u003eLiu, M., Zhang, F., Huang, H. H. (2017). An adaptive classification strategy for reliable locomotion mode recognition. Sensors (Basel). 17. doi: 10.3390/s17092020.\u003c/li\u003e\n\u003cli\u003eMartinez-Villalpando, E. C., Herr, H. (2009). Agonist-antagonist active knee prosthesis: A preliminary study in level-ground walking. J. Rehabil. Res. Dev. 46, 361\u0026ndash;373.\u003c/li\u003e\n\u003cli\u003eMohd Yusof, N. H. Mohd., Hamzaid, N. A., Jasni, F., Lai, K. W. (2018). In-socket sensory system with an adaptive neuro-based fuzzy inference system for active transfemoral prosthetic legs. J. Electron. Imag. 28, 1. doi: 10.1117/1.JEI.28.2.021002.\u003c/li\u003e\n\u003cli\u003ePfeifer, S., Pagel, A., Riener, R., Vallery, H. (2015). Actuator with angle-dependent elasticity for biomimetic transfemoral prostheses. IEEE/ASME Trans. Mechatron. 20, 1384\u0026ndash;1394. doi: 10.1109/TMECH.2014.2337514.\u003c/li\u003e\n\u003cli\u003ePieringer, D. S., Grimmer, M., Russold, M. F., Riener, R. (2017). Review of the actuators of active knee prostheses and their target design outputs for activities of daily living. IEEE Int. Conf. Rehabil. Robot.. 2017, 1246\u0026ndash;1253. doi: 10.1109/ICORR.2017.8009420.\u003c/li\u003e\n\u003cli\u003eSaglam, Y., Gulenc, B., Birisik, F., Ersen, A., Yilmaz Yalcinkaya, E., Yazicioglu, O. (2017). The quality-of-life analysis of knee prosthesis with complete microprocessor control in trans-femoral amputees. Acta Orthop. Traumatol. Turc. 51, 466\u0026ndash;469. doi: 10.1016/j.aott.2017.10.009.\u003c/li\u003e\n\u003cli\u003eSegal, A. D., Orendurff, M. S., Klute, G. K., McDowell, M. L., Pecoraro, J. A., Shofer, J. et al. (2006). Kinematic and kinetic comparisons of transfemoral amputee gait using C-Leg and Mauch SNS prosthetic knees. J. Rehabil. Res. Dev. 43, 857\u0026ndash;870. doi: 10.1682/jrrd.2005.09.0147.\u003c/li\u003e\n\u003cli\u003eShaikh, S., Malhotra, A. (2020). Real-time feedback control for knee prosthesis using motion fusion algorithm in 6-dof IMU. J. Sci. Ind. Res. (India). 79, 213\u0026ndash;215.\u003c/li\u003e\n\u003cli\u003eSong, L., Wang, X., Gong, S., Shi, Z., Chen, L. (2008). Design of active artificial knee joint. IFMBE Proc. 19, 155\u0026ndash;158. doi: 10.1007/978-3-540-79039-6_40.\u003c/li\u003e\n\u003cli\u003eSup, F., Bohara, A., Goldfarb, M. (2007a). Design and control of a powered knee and ankle prosthesis, in Proc. - IEEE International Conference on Robotics and Automation, 4134\u0026ndash;4139.\u003c/li\u003e\n\u003cli\u003eSup, F., Bohara, A., Goldfarb, M. (2007b). Design and control of a powered knee and ankle prosthesis. Proc. \u0026ndash; IEEE Int. Conf. Robot. Autom., no. April, 4134\u0026ndash;4139.\u003c/li\u003e\n\u003cli\u003eSup, F., Varol, H. A., Mitchell, J., Withrow, T., Goldfarb, M. (2008a). Design and control of an active electrical knee and ankle prosthesis. Proc. 2nd Bienn. IEEE/RAS-EMBS Int. Conf. Biomed. Robot. Biomechatronics, Scottsdale, AZ, USA, 523\u0026ndash;528.\u003c/li\u003e\n\u003cli\u003eSup, F., Varol, H. A., Mitchell, J., Withrow, T., Goldfarb, M. (2008b). Design and control of an active electrical knee and ankle prosthesis, in Proc. 2nd Biennial IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics. Proc. IEEE RAS EMBS Int. Conf. Biomed. Robot. Biomechatron. 2008, 523\u0026ndash;528. doi: 10.1109/BIOROB.2008.4762811.\u003c/li\u003e\n\u003cli\u003eThatte, N., Geyer, H. (2016). Toward balance recovery with leg prostheses using neuromuscular model control. IEEE Trans. Biomed. Eng. 63, 904\u0026ndash;913. doi: 10.1109/TBME.2015.2472533.\u003c/li\u003e\n\u003cli\u003eThiele, J., Westebbe, B., Bellmann, M., Kraft, M. (2014). Designs and performance of microprocessor-controlled knee joints. Biomed. Tech. 59, 65\u0026ndash;77.\u003c/li\u003e\n\u003cli\u003eTorrealba, R. R., Fern\u0026aacute;ndez-L\u0026oacute;pez, G., Grieco, J. C. (2008). Towards the development of knee prostheses: Review of current researches. Kybernetes. 37, 1561\u0026ndash;1576.\u003c/li\u003e\n\u003cli\u003eVarol, H. A., Goldfarb, M. (2007a). Decomposition-based control for a powered knee and ankle transfemoral prosthesis 10th Int. Conf. Rehabil. Robot. ICORR\u0026rsquo;07, 00 (IEEE Publications), 783\u0026ndash;789.\u003c/li\u003e\n\u003cli\u003eVarol, H. A., Goldfarb, M. (2007b). Decomposition-based control for a powered knee and ankle transfemoral prosthesis 10th Int. Conf. Rehabil. Robot. ICORR\u0026rsquo;07, 00, no. c, (IEEE Publications), 783\u0026ndash;789.\u003c/li\u003e\n\u003cli\u003eVarol, H. A., Sup, F., Goldfarb, M. (2009). Real-time gait mode intent recognition of a powered knee and ankle prosthesis for standing and walking. Proc. IEEE RAS EMBS Int. Conf. Biomed. Robot Biomechatron. 2008, 66\u0026ndash;72. doi: 10.1109/BIOROB.2008.4762860. \u003c/li\u003e\n\u003cli\u003eVarol, H. A., Sup, F., Goldfarb, M. (2010). Multiclass real-time intent recognition of a powered lower limb prosthesis. IEEE Trans. Biomed. Eng. 57, 542\u0026ndash;551. doi: 10.1109/TBME.2009.2034734.\u003c/li\u003e\n\u003cli\u003eWen, Y., Si, J., Brandt, A., Gao, X., Huang, H. H. (2020). Online reinforcement learning control for the personalization of a robotic knee prosthesis. IEEE Trans. Cybern. 50, 2346\u0026ndash;2356. doi: 10.1109/TCYB.2019.2890974.\u003c/li\u003e\n\u003cli\u003eWu, R., Li, M., Yao, Z., Si, J., He, Huang. (2021). Reinforcement learning enabled automatic impedance control of a robotic knee prosthesis to mimic the intact knee motion in a co-adapting environment.\u003c/li\u003e\n\u003cli\u003eWu, S. K., Waycaster, G., Shen, X. (2011). Electromyography-based control of active above-knee prostheses. Control Eng. Pract. 19, 875\u0026ndash;882. doi: 10.1016/j.conengprac.2011.04.017.\u003c/li\u003e\n\u003cli\u003eYoung, A. J., Kuiken, T. A., Hargrove, L. J. (2014). Analysis of using EMG and mechanical sensors to enhance intent recognition in powered lower limb prostheses. J. Neural Eng. 11, 056021. doi: 10.1088/1741-2560/11/5/056021.\u003c/li\u003e\n\u003cli\u003eYoung, A. J., Simon, A. M., Hargrove, L. J. (2014). A training method for locomotion mode prediction using powered lower limb prostheses. IEEE Trans. Neural Syst. Rehabil. Eng. 22, 671\u0026ndash;677. doi: 10.1109/TNSRE.2013.2285101.\u003c/li\u003e\n\u003cli\u003eYoung, A. J., Simon, A., Hargrove, L. J. (2013). An intent recognition strategy for transfemoral amputee ambulation across different locomotion modes. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS. 2013, 1587\u0026ndash;1590. doi: 10.1109/EMBC.2013.6609818.\u003c/li\u003e\n\u003cli\u003eZlatnik, D. (1998). Intelligently controlled above knee (A/K) prosthesis \u003cem\u003eInst. Robot. ETH- Zurich. Switz.\u003c/em\u003e \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"physical-and-engineering-sciences-in-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"apes","sideBox":"Learn more about [Physical and Engineering Sciences in Medicine](http://link.springer.com/journal/13246)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/apes/default.aspx","title":"Physical and Engineering Sciences in Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"active prosthetic device, user-intention, control algorithm, sensory system, intent recognition","lastPublishedDoi":"10.21203/rs.3.rs-2814842/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2814842/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn this paper, a cutting-edge method for creating control interfaces for intention detection in active transfemoral prosthetic devices is presented. Given the current trend in the prosthetics industry, a review of the literature over the last two decades has found a number of control algorithms utilized for intention detection implementation. Scientific publications, books, and online resources were evaluated for published material on knee prosthesis. Based on the materials, three areas of scientific inquiry involving control interfaces for intention detection in active prosthetic legs have been identified. The studies were assessed using the Downs and Black checklist, and their control techniques as well as the performance assessment for these control interfaces are described. After screening, 211 studies were retrieved and examined, however only 39 publications were included and examined in this review. An active prosthetic leg's control strategy framework and goal output were examined in fifteen (15) papers. In two (2) papers, conventional control methods for transfemoral prosthetic legs were examined. Eight (8) further research looked at the potential implementation of intent detection in the transfemoral prosthetic leg, while fourteen (14) papers explored the active prosthetic leg's machine learning algorithm. As a result, our research showed that using a less complex sensory system to control an active transfemoral prosthetic limb is possible when paired with a creative approach and control algorithm that can translate the restricted sensor data into a larger set of relevant data. Therefore, an effective sensory system (practicality and quality of the sensory input) and intention detection algorithm were required for an active transfemoral prosthetic limb. It is considered a complex activity because of the connected biomechanics of human gait, which the brain governs unconsciously. No prosthetic device has been able to replace a human limb in the same way that the original one would operate due to the ongoing challenges in prosthetic technology. Even with something as simple as walking, there are several components to the difficulty that may be broken down into different study areas. No matter how little, this calls for input from many different domains of knowledge. Considerations for future development might be based on this current analysis of the intention-detecting control interface of active knee prosthetic devices and prosthetic technology. This evaluation of the literature includes not only a study of the viability of control interfaces that support intention detection for an active prosthetic limb, but it also suggests a framework for categorizing various works in the subject.\u003c/p\u003e","manuscriptTitle":"Control Interfaces for Intention Detection in Active Transfemoral Prosthetics: A Systematic Review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-04-20 13:36:13","doi":"10.21203/rs.3.rs-2814842/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revisions","date":"2023-11-23T06:58:02+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2023-06-12T15:42:30+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2023-04-18T15:06:24+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Physical and Engineering Sciences in Medicine","date":"2023-04-16T23:57:58+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-04-15T07:44:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"Physical and Engineering Sciences in Medicine","date":"2023-04-13T20:59:14+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"physical-and-engineering-sciences-in-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"apes","sideBox":"Learn more about [Physical and Engineering Sciences in Medicine](http://link.springer.com/journal/13246)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/apes/default.aspx","title":"Physical and Engineering Sciences in Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"438375a4-1b71-4a0a-84a3-b8fe4931fcb6","owner":[],"postedDate":"April 20th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-11-03T16:03:23+00:00","versionOfRecord":{"articleIdentity":"rs-2814842","link":"https://doi.org/10.1007/s13246-025-01623-0","journal":{"identity":"physical-and-engineering-sciences-in-medicine","isVorOnly":false,"title":"Physical and Engineering Sciences in Medicine"},"publishedOn":"2025-10-31 15:58:39","publishedOnDateReadable":"October 31st, 2025"},"versionCreatedAt":"2023-04-20 13:36:13","video":"","vorDoi":"10.1007/s13246-025-01623-0","vorDoiUrl":"https://doi.org/10.1007/s13246-025-01623-0","workflowStages":[]},"version":"v1","identity":"rs-2814842","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2814842","identity":"rs-2814842","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}