New Possibilities of A Quantitative Assessment Method for the Jebsen-Taylor Hand Function Test: A Preliminary Study

New Possibilities of A Quantitative Assessment Method for the Jebsen-Taylor Hand Function Test: A Preliminary Study | 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 Article New Possibilities of A Quantitative Assessment Method for the Jebsen-Taylor Hand Function Test: A Preliminary Study Na-Yun Seo, Joo-Hyun Lee, Young-Jin Jung This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4464229/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The Jebsen–Taylor Hand Function Test (JTHFT) is a standardized and objective assessment tool for evaluating hand dysfunction in various conditions (including stroke and spinal cord injury). The JTHFT has been widely used successfully in clinical settings; however, the instrument score does not reflect mechanical movement information of the upper extremities. Therefore, we developed a software to quantitatively evaluate the JTHFT. The subject’s motion was recorded using a webcam and a deep learning-based three-dimensional motion-extraction technique. Several elements were calculated from vectors between two points. Consequently, eight factors were analyzed: distance to move beans, velocity of the hand moving beans, time to move beans, time to put beans in a spoon, time to put beans in a cup, angle of the dorsum of the hand, trajectory of the hand, and total task time. The average task time was 6.82 s (standard deviation, 0.09), which within the normal range for adults. The cross-product angle of the dorsum of the hand was confirmed to be 0.01 units, depending on how tilted the hand was. In summary, we identified new quantitative assessment tools for occupational therapy; the extracted index results allowed for quantitative evaluation of the aspect that was previously impossible to judge. Health sciences/Health occupations Physical sciences/Engineering Physical sciences/Nanoscience and technology Deep Learning Computer-aided Rehabilitation Hand Function Motion Trajectory Quantitative Assessment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION The Jebsen–Taylor Hand Function test (JTHFT) was developed in 1969 to offer an objective and standardized assessment of activities of daily living [ 1 ]. The JTHFT is generally used in occupational therapy for patients with various conditions associated with brain injury and is mainly used to evaluate the function of patients with hemiplegia after stroke [ 2 ]. The JTHFT only uses time as a parameter. In other words, the JTHFT focuses on time rather than task performance. Seven tasks are included in the JTHFT: writing a sentence, card turning, lifting small objects, simulated feeding, stacking checkers, picking up light cans, and picking up heavy cans [ 3 ]. Occupational therapy can help patients regain independence in activities of daily living [ 4 ]. Among several assessments, the JTHFT is mainly used in hospitals to evaluate hand function, and its reliability has been proven in several studies [ 5 ]. In 2015, Mak et al. [ 2 ] compared the difference between patients with Parkinson’s disease and healthy individuals using the JTHFT. Thus, the JTHFT was judged to be a reliable tool that can evaluate the hand function of patients with Parkinson’s disease. In 2020, Aynaci et al. [ 6 ] reported that objective measures, such as the JTHFT, could help clinicians obtain medical evidence from patients. In 2018, Lim et al. [ 7 ] measured the reliability of the JTHFT using standard translated instructions and spontaneously translated instructions, reporting its usefulness. In 2021, Lin et al. [ 8 ] measured the reliability and responsiveness of the JTHFT and BBT in children with cerebral palsy. Both tests showed high reliability when performed within a short time. In 2021, Sığırtmaç et al. [ 1 ] proved that JTHFT is a reliable and valid tool and reported that clinicians and researchers could use it to evaluate patients' hand function. JTHFT can evaluate improved hand function after occupational therapy; however, occupational therapists have identified the following limitations. The results of the JTHFT are only indicated as the time for assessing the agility of the hand. However, this variable cannot determine complex upper-extremity movement patterns such as smoothness, velocity, and coordination during performance assessment. In addition, the change (improvement) in the effects of occupational therapy is too slow to be judged with the naked eye. This can limit a patient's motivation for treatment. In occupational therapy, the patient's passion for participating in voluntary treatment plays an important role. A study revealed a significant difference in functional recovery among patients who had an active willingness to undergo treatment compared to patients who underwent passive treatment. For example, Yoshida et al. [ 9 ] revealed that patients are greatly motivated when they recognize a change in their body state during the treatment process. Occupational therapy is an essential step for patients, which can relieve many of the inconveniences they experience in daily life and improve their quality of life [ 10 ]. To motivate patients and restore their ability to perform tasks, occupational therapists have included various compensation strategies for patients during the treatment process [ 11 ]. Occupational therapists must perform several tasks. Occupational therapy after a stroke is time-consuming for occupational therapists and can incur high medical costs or work burdens. Professionalism is the main reason for this increase in medical costs. Particularly, high professionalism is required for therapists to visually judge the degree of the effect of occupational therapy on the patient. To obtain high expertise as occupational therapists, years of patient treatment experience and intensive education are required. This process incurs high costs for training therapists, leading to an increase in medical costs. Consequently, the cost of medical services borne by the patients increases. The high cost of medical services reduces the number of times patients visit the hospital, which reduces the effectiveness of occupational therapy for patients and can decrease their quality of life. In addition to the treatment cost, there is an occupational therapist’s work burden, which reduces the quality of life. A high workload in the treatment process causes fatigue in therapists, which can lower the quality of medical services [ 12 ]. This can negatively affect patients and decrease the treatment effects. To overcome the abovementioned limitations and workload of therapists, various techniques for evaluating hand function are being studied. Yin et al. [ 13 ] automatically evaluated the hand functions using wearable gloves with multiple sensors. Three inertial measurement unit (IMU) sensors and one thin pressure sensor were used to capture the hand movement, and the pressure sensor measured the fingertip pressure. Using this method allowed for the automatic evaluation of the range of motion, flexibility, and scores. Kim et al. [ 14 ] developed a wearable hand exoskeleton system to analyze the finger muscle forces. The system calculated the load on the fingertips using cable-driven actuators. Zimmerman et al. [ 15 ] measured finger flexion using a glove equipped with an analog flex sensor and measured the position and orientation of the hand using ultrasonic and magnetic flux sensors. Zheng et al. [ 16 ] used sensor gloves to quantitatively measure hand movements. Two types of sensors were used to measure the angle of the finger joint and force exerted by the finger on the object, and their validity and reliability were evaluated. Lin et al. [ 8 ] developed data gloves for the rehabilitation and assessment of hand function. An IMU sensor and a sensor-fusion algorithm were used in the glove. Its modular design enables more data collection with high scalability. The leaf motion controller (LMC) by Houston et al. [ 17 ] provided a low-cost method with markerless hand tracking. Several LMCs were used on a single computer to measure hand motions in multiple directions. The Kalman filter and Kabsch algorithm were used, and their reliability and validity were evaluated by comparing the CMC and MJA. Kawaguchi et al. [ 18 ] measured hand movements using an inexpensive and easy-to-use 3D range image sensor. To evaluate the actual measurement accuracy, the error was calculated using measurements from motion capture. The Vicon motion-capture system was used by Kontson et al. [ 19 ]. The average range of motion and maximum angle of the JTHFT task, performed by attaching a marker to the subjects' bodies, was calculated, and upper-body joint data were collected. Kapuscinski et al. [ 20 ] constructed a gesture-based interface and used skin-colored regions to estimate the hand position. After handshape recognition, gestures were recognized using the hidden Markov model. Yu et al. [ 21 ] developed a system for extracting hand-gesture features based on a multilayered perceptron. A total of 3,500 images were used as the dataset, and a system that can recognize hand motion through the Hu-invariant moment and Fourier descriptor was proposed. Malima et al. [ 22 ] proposed an algorithm for recognizing motions in hand images. This system comprises three steps: hand region segmentation, finger segmentation, and gesture classification. Koh et al. [ 23 ] proposed a vision-based method that can extract the hands without being limited by lighting conditions. To effectively recognize hands in the system, a method for modeling skin color after distinguishing hands from objects was used. These technologies exhibit good performance in evaluating hand function; however, they are insufficient to be used as quantitative evaluation tools instead of the JTHFT by occupational therapists. Although the vision-based method provides accurate positional information with an error of < 1 mm, it requires a complex computational process after motion data acquisition. If a quantitative evaluation tool is developed, it can reduce the burden on the therapist and the medical service cost for increasing the quality of life of patients. Therefore, we examined the possibility of a new quantitative evaluation technique to overcome the aforementioned limitations of JTHFT analysis. In this study, video data were recorded using a webcam. As the motion was recorded using a webcam, the subject could perform the task without interference. We developed custom-made SW software to analyze the video data. SW recognizes the human hand and restores the finger joint from a 2D image to 3D data. Based on the joint position information, various features, such as distance, velocity, time, angle, and trajectory, were extracted. This analysis method uses videos. Therefore, video-based evaluations were possible. Consequently, the evaluation data were preserved. This allows occupational therapists to make evidence-based diagnoses. In addition, homemade SW derives quantitative and objective evaluations. Therefore, patients can judge the extent to which their function has improved. These advantages can reduce the workload of the therapist and motivate patients for treatment. In this paper, “Simulated Feeding,” which is one of the JTHFT tasks, was analyzed. “Simulated Feeding” is a task that is most closely related to daily life. In this study, we attempted to confirm the possibility of a quantitative evaluation applicable to occupational therapy assessment. METHODS This study was approved by the ethics committee of the Institutional Review Board of Baekseok University (BUIRB-202307-HR-020). The study was conducted according to the guidelines of the Declaration of Helsinki, and written informed consent was obtained from the subject before participation. Experiment set-up A laptop (Legion Y720, Lenovo, China) and a webcam (C920 PRO HD webcam, Logitech, Switzerland) were used to record video images. MATLAB (MATLAB R2022a, MathWorks, TX, USA) was used for program analysis. The recorded video had a resolution of 1280 × 720 pixels, and the frame rate was 30 frames per second. When the focus was set to auto, the screen often zoomed in and blurred. Therefore, the focus was set to 1% rather than auto. The homemade application was designed to estimate the quantitative index automatically measured for analysis. To save a video, the name was set using only English and underbar. In this experiment, a webcam was installed on a tripod at a height of 90 cm. The webcam was tilted at a 60-degree. The lens was pointed downwards to confirm the change in distance. The diameter of the beans used in the experiment was 1.6 cm, and the spacing between the beans was 5 cm. Five beans were set side-by-side in a straight line. The beans were set approximately 30 cm away from the subject, and the cup was set approximately 10 cm away from the 1st bean. The right hand was placed approximately 10 cm from the 5th bean. The total distance from the cup to the hand was approximately 60 cm. The hand holding the spoon was 10 cm in diameter (Fig. 1 ). Data from the same subject performing the task a total of 30 times were used for the analysis. The average video running time was 11.4 s, and the average video size was 26,435 KB. A CSV file containing joint coordinates was used for the analysis. The CSV file contained data extracted from the AI-Motion app using the MediaPipe core. The CSV file contained the values of the X-, Y-, and Z-coordinates corresponding to each joint. There were 21 landmarks for the hand. Therefore, in this file, the number of columns was 63, and the number of rows was the number of video frames. Twenty-one landmarks on the hand were used. Experiment task (Task selection) Seven tasks are included in JTHFT. Simulated Feeding is the fourth task in the JTHFT. In this study, a Simulated Feeding task that is most closely related to daily life was selected and analyzed. Simulated Feeding involves transferring beans to a cup with a spoon. The experimental method is as follows: (For easy explanation, the bean number on the far left is set to 1, and the bean number on the far right is set to 5) The subject grabs the spoon and places his hand 10 cm from the fifth bean. When the video recording starts, the subject waits for about 2 s without moving. After 2 s, the subject places the fifth bean in the spoon and transfers it to a cup. Subsequently, the 4th, 3rd, 2nd, and 1st beans are transferred to the cup. After moving all five beans, the subject returns the spoon to the initial position as the starting point. After returning to the initial position, the subject waits for about 2 s until the video finishes recording. During this task, the patient’s fingers, dorsum of the hand, and wrist were recorded to estimate the 3D joint position from the 2D webcam image. Evaluation procedure This study used homemade software to quantitatively evaluate the Simulated Feeding, which was previously qualitatively assessed. This software was developed for the convenience of occupational therapists and can be used without engineering knowledge. The joint points of the hand were extracted using the core of the Mediapipe in the software. Factors that could be analyzed through a Simulated Feeding task, such as the distance, velocity, time, angle, and trajectory, were calculated using this software. The SW processing is illustrated in Fig. 3 . The motion was recorded using a webcam connected to a laptop. The joint points were used as inputs for the task analysis model to calculate the factors. Finally, the analyzed elements were output as a PDF (Fig. 2 (a)). Distance calculation To calculate the distance of the task, we used coordinate 12, as illustrated in Fig. 2 (b). The MIDDLE_FINGER_TIP was used as the reference point. This is because the joint has less movement than the other finger joints while holding a spoon. The Euclidean distance formula was used to calculate the distance between two points. The calculated distances are the relative distance values. Thus, we used the ratio of the relative value with the actual value to calculate the actual distance. In this ratio, the relative value is the distance between the first and last points of the task derived from the analysis. The actual value is the distance between the starting point and the cup. This was measured using a tape measure. The actual distance was calculated using two ratios. Figure 3 (a) illustrates the calculated distance as a graph. The “diff function” was used to calculate the acceleration values. Acceleration values were used to distinguish the tasks, as illustrated in Fig. 3 (a). Subsequently, acceleration was classified using a pulse period function. The pulse period function determines the start and end points of the signal. After applying the “diff function” and “pulseperiod function” to Fig. 4 , the signal outputs as illustrated in Fig. 3 (b). $$Distance= \sqrt{{({p}_{1}-{q}_{1})}^{2}+{({p}_{2}-{q}_{2})}^{2}+{({p}_{3}-{q}_{3})}^{2}}$$ 1 Calculation of distance, velocity, and time for each task Task classification was calculated using the orange pillars shown in Fig. 3 (b). The number of pillars was 11. Therefore, the pillar had a 2 × 11 matrix value (Pa matrix), which comprised the start and endpoints. The remaining data, except for those in the first column, are a 2 × 10 matrix (= Pb matrix). It contains the start and end data for the task. In this calculation, Pa and Pb matrices were used. A Pa matrix was used to calculate the starting point of the subject’s motion. A Pb matrix was used to calculate the values for each task. Simulated Feeding is a task in which the subject transfers beans and returns five times. The beans were transferred to the cup in the “transferring bean section.” These sections correspond to the 1st, 3rd, 5th, 7th, and 9th columns of the Pb matrix, respectively. In other words, these sections can form a 2 × 5 matrix (Po matrix), which is an odd column in the Pb matrix. Similarly, the “return section” is the returning section after the transferring bean section. These sections correspond to the 2nd, 4th, 6th, 8th, and 10th columns of the Pb matrix, respectively. These sections can form a 2 × 5 matrix (Pe matrix), which is an even column in the Pb matrix. Each task involved transfer and return motions. This means that each pillar has odd and even values. With this value, the transferring bean section and return section can be calculated. For example, to calculate the distance of the 1st bean from the cup (1st transferring bean section), the first column of the odd array was used. The distance can be calculated by subtracting Ps(1,1) from Ps(2,1). At this point, the matrix value should be substituted with the distance value calculated in Section 3.1. Therefore, a substituted value was used for subtraction. The calculation of the distance for the “return section” is the same as above. However, in this study, data from the Pe matrix were used. The time required per task was calculated as the number of frames per 30 iterations. We divided the video by 30 because the frame rate was 30 frames per second. The formula for calculating the velocity was distance/time. The velocity corresponding to each task was calculated using the distance per task/time per task formula. Cross product angle of the dorsum of the hand The cross product was obtained from the dorsum of the hand. The cross-product is a vector pointing upward at 90 ° from the dorsum of the hand. There are three areas on the dorsum of the hand: 0-5-9, 0-9-13, and 0-13-17. These areas are created by connecting the landmarks in Fig. 2 (b). The cross product of each area was obtained from these three areas. The average cross-product values for the three areas were calculated. The cross product of the first 10 frames was used to set the reference point. As a result, the cross product was calculated using Eq. 3 . $$\overrightarrow{a} \bullet \overrightarrow{b}= \left|\overrightarrow{a}\right| \left|\overrightarrow{b}\right| cos\theta$$ 2 $$cos\theta = \frac{\overrightarrow{a} \bullet \overrightarrow{b}}{\left|\overrightarrow{a}\right| \left|\overrightarrow{b}\right|}$$ 3 $$\left|\overrightarrow{a}\right|= \sqrt{{V}_{x}^{2}+{V}_{y}^{2}+ {V}_{z}^{2}}$$ 4 $$\overrightarrow{a} \bullet \overrightarrow{b}= {a}_{x}{b}_{x}+ {a}_{y}{b}_{y}+ {a}_{z}{b}_{z}$$ 5 Trajectory The trajectory represents hand movements for each task. In this calculation, the Pb matrix in 3.2 was also used. Afterward, the “scatter3 function” and “colormap” were used. These functions show that the color of the dots changes according to the changes in hand movement. RESULTS Distance In Fig. 3 (a), the part where the graph rises is the “transferring bean section.” Conversely, the “return section” is the section where the graph decreases. The graph length of the “transferring bean section” became shorter with time. This was because the subject's hand gradually moved from the 5th bean to the 1st bean. BCD is the distance of the “transferring bean section,” implying that BCD is the distance from bean to cup. RD is the distance of the “return section,” implying that RD is the distance from cup to bean. “S” is the section where beans are put into a spoon. In Fig. 3 (b), the line arrow ascending is a section of BCD, and the dotted line arrow descending is the section of RD. The small rectangle is a section of S. The combination of the line and dotted pillars is a Pa matrix. Line pillars represent the Pb matrix. The X-axis represents the number of video frames, and the Y-axis represents distance data. Table 1 presents the results obtained by performing the task 30 times by the same person. The values ​​in Table 1 gradually decrease during the task. Theoretically, the values of the RD (Task 1) and BCD (Task 2) should be identical. This is because the actions occur at the same location. However, the two values ​​ differed. This occurred because the subject performing the task is a human and not a machine. In other words, the subjects moved their hands while trying to put the beans into the spoon. There was no significant difference in the values ​​because the behavioral ranges were short. For example, in Table 1 , the difference between the RD (Task 1) and BCD (Task 2) was approximately 4 cm. The difference between the RD (Task 2) and BCD (Task 3) was approximately 2 cm. However, in the case of RD, the value for Task 5 increased rapidly. This was because the position of the hand changed. Tasks 1 to 4 involved moving from cup to bean. However, Task 5 involved returning from the cup to the first hand position. Therefore, because task 5 involves returning from the cup to the hand’s position where the task first starts, the distance value increases rapidly. Table 1 Distance of moving bean to cup and return distance Variable Bean to Cup Distance (BCD) Return Distance (RD) Mean (m) STD (m) Mean (m) STD (m) Task 1 0.35 0.04 0.25 0.02 Task 2 0.29 0.02 0.21 0.01 Task 3 0.23 0.02 0.14 0.01 Task 4 0.17 0.01 0.09 0.01 Task 5 0.12 0.01 0.51 0.02 Tables 2 and 3 present the velocity and time of moving the bean to the cup and returning, respectively. The velocity is a relative value. Table 3 shows that the greater the number of tasks, the shorter the time. This was possibly because the subjects became more accustomed to performing the task, shortening the time. Table 2 Velocity of moving bean to cup and return Variable Bean to Cup Velocity (BCV) Return Velocity (RV) Mean STD Mean STD Task 1 0.71 0.09 0.75 0.11 Task 2 0.69 0.08 0.62 0.08 Task 3 0.63 0.08 0.58 0.08 Task 4 0.59 0.07 0.50 0.11 Task 5 0.52 0.08 1.03 0.16 Table 3 Time of moving bean to cup and return Variable Bean to Cup Time (BCT) Return Time (RT) Mean (s) STD (s) Mean (s) STD (s) Task 1 0.49 0.07 0.34 0.08 Task 2 0.43 0.09 0.34 0.07 Task 3 0.37 0.07 0.26 0.06 Task 4 0.29 0.05 0.19 0.07 Task 5 0.23 0.05 0.51 0.09 Cross product angle of the dorsum of the hand Figure 4 includes five parts in which the value of the graph increases. This part turns the wrist when the beans are placed in the cup. By specifying the task execution range, the maximum cross-product angle was used to quantify the wrist rotation when performing the task. In Table 4 , the angle was calculated in units of 0.01°. The closer the value of the graph is to 90 °, the more the wrist rotates by approximately 90 °. Through this maximum cross-product angle, the extent to which the subjects rotate their wrist when placing the bean into the cup could be determined. The X-axis represents the number of video frames, and the Y-axis represents the angle data. Table 4 Cross product angle of the dorsum of the hand Variable Cross product angle of the dorsum of the hand ( \(^\circ\) ) Task 1 82.85 Task 2 82.80 Task 3 59.27 Task 4 82.47 Task 5 43.66 Trajectory Figure 5 shows the hand movement trajectory for each task. The blue trajectory is the “transferring bean section.” The green trajectory is the section where the beans were placed in the cup. The red trajectory represents the “return section.” Five trajectories were extracted because there were five tasks. The shape of the 5th trajectory differed from those of the other four trajectories. For the 5th trajectory, the red trajectory was longer than the blue one. The 5th trajectory is the section where the 1st bean was moved to the cup and returned. The blue trajectory represents the distance from 1st bean to the cup. The red trajectory represents the distance from the cup to the first hand position. The first hand position was next to the 5th bean. The total length of the trajectory is listed in Table 5 . Thus, we could calculate the total trajectory length. The values are relative. Based on the calculated value, we could determine whether the subject performed an unnecessary motion. The X-axis represents the number of video frames, and the Y-axis represents the angle data. Table 5 Relative length of the trajectory Variable Relative length of the trajectory Task 1 0.53 Task 2 0.40 Task 3 0.32 Task 4 0.27 Task 5 0.46 Time Table 6 lists these two types of times. The first is the time required to place beans in the cup. The second was the time required to place beans in the spoon (BST). The average BCT was approximately 0.174 s. The average BST was approximately 0.438 s. This indicates that the subjects spent more time putting the beans in the spoon than putting it in the cup. Table 6 Time of putting bean into the cup and into the spoon Variable Put Bean into Cup Time (PBCT) Bean to Spoon Time (BST) Mean (s) STD (s) Mean (s) STD (s) Task 1 0.16 0.05 0.26 0.30 Task 2 0.17 0.05 0.45 0.20 Task 3 0.16 0.05 0.42 0.22 Task 4 0.18 0.06 0.48 0.07 Task 5 0.20 0.10 0.58 0.11 In Table 7 , the total time for the task is the time used only for the task (the waiting time of approximately 2 s at the start and end of the task in the experimental method was excluded). In this study, the derived time is 6.82 s. In the case of normal adults (20–59 s), the time average value is approximately 7.1 ± 1.4 s. This value is the standardized time of Simulated Feeding. The time derived in this study includes the normal value range. Table 7 Total task time and normal time Variable Mean (s) STD (s) Normal (s) Total task time 6.82 0.09 7.1 ± 1.4 DISCUSSION The method proposed in this paper quantitatively calculates the elements of the Simulated Feeding of JTHFT using a webcam. The conventional Simulated Feeding was judged only by time. During occupational therapy, the shape of the subject’s motion or the angle of hand tilting is impossible to calculate. Moreover, quantitative analysis is not possible. Therefore, it is difficult for patients to confirm that their body function has improved. These limitations make it difficult to motivate patients and can place a burden on occupational therapists. To overcome these limitations, several techniques have been studied to quantitatively evaluate hand function. These techniques can be used to quantitatively evaluate hand function; however, they have certain limitations. For the wearable device, a glove is worn on the hand. This has the disadvantage that external stimuli exist during the performance of the JTFHT. Therefore, patients may not be able to use their full strength during the task. Although low-cost, easy wearability, and high-reliability wearable devices are available, it is difficult for patients with hand discomfort to wear them. In the case of the marker-based method, the marker is attached to the body, and it is inconvenient for patients to perform these tasks. In addition, the markers are expensive. Among the sensors, some have a low cost, such as the LMC. However, this sensor has a narrow field of view and is highly dependent on the sensor position. Basically, this sensor has space constraints. Finally, in the case of most vision-based methods, the results vary depending on lighting because the hand is recognized based on the skin color. To overcome these limitations, we quantitatively analyzed the JTHFT using a homemade SW. As a webcam was used, it was not necessary to attach it to the patient's body. Moreover, they are inexpensive and have no space constraints. In addition, because the SW uses the coordinates of the hand, it is less affected by lighting than conventional vision methods. Using the homemade SW, we quantitatively calculated the elements of JTHFT: distance, velocity, angle, and time. In addition, this SW can schematize the hand's trajectory and the cross-product of the dorsum of the hand, which is otherwise impossible to measure. However, because this study used a single webcam, several issues must be addressed. Recording from top to bottom makes it easy to calculate the distance. However, it is impossible to calculate depth information. In addition, it is difficult to recognize these coordinates. Coordinates were sometimes lost when all fingers were not visible on the webcam screen. However, this issue can be resolved by changing from a single to multiple webcams. If these limitations are resolved, this method can be widely used in occupational therapy, where quantitative judgment is difficult, and in other fields of rehabilitation. Conclusions In this paper, the homemade SW quantitatively analyzed Simulated Feeding of JTHFT. The distance, velocity, and time the beans moved were calculated using this SW. In addition, we could calculate the angles of the fingers and the cross-product of the dorsum of the hand, which changed while performing the task. Using a schematic of the trajectory, we determined whether there were any unnecessary movements while the subject performed the task. We confirmed that the total task time derived from SW was included in the average Simulated Feeding time. Thus, we can conclude that the developed SW quantitatively analyzed the Simulated Feeding. We believe that this SW can be used worldwide in areas that require quantitative calculations. Declarations Competing interest statement : The authors declare that they have no competing interests. Author Contribution Young-Jin Jung and Joo-Hun Lee conceived the original idea. Na-Yun Seo wrote the main manuscript text and Young-Jin Jung prepared study design and settings. This was also discussed and reviewed with all authors. Acknowledgement This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2022R1C1C1010127) Data Availability The datasets used and analyzed in the current study are available from the corresponding author upon reasonable request. References Sığırtmaç, İ. C. & Öksüz, Ç. Investigation of reliability, validity, and cutoff value of the Jebsen-Taylor Hand Function Test. J. Hand Ther. 34, 396–403 (2021). Mak, M. K. Y., Lau, E. T. L., Tam, V. W. K., Woo, C. W. Y. & Yuen, S. K. Y. Use of Jebsen Taylor Hand Function Test in evaluating the hand dexterity in people with Parkinson’s disease. J. Hand Ther. 28, 389–394; quiz 395 (2015). Ladha, N. & Gaidhani, G. Assessment of hand function in poststroke patients using ‘Jebsen Taylor Hand function test’. Indian J. Physiother. Occup. 15, 97–101 (2021). Kwakkel, G., et al . Effects of augmented exercise therapy time after stroke: A meta-analysis. Stroke 35, 2529–2539 (2004). Sears, E. D. & Chung, K. C. Validity and responsiveness of the Jebsen-Taylor hand function test. J. Hand Surg. Am 35, 30–37 (2010). Aynaci, G. & Kaya, B. Evaluation of Jebsen-Taylor Hand Function Test for Use in Nursing Students: Close-Future Outlook . International Balkan Education and Science Congress. 13, 711–718 (2018) Lim, Y. X. & Chai, S. C. Standardized translated instruction versus spontaneously translated instruction: test-retest and interrater reliability of a hand function test. J. Hand Ther. 33, 553–561 (2020). Lin, B. S. et al. Design of an inertial-sensor-based data glove for hand function evaluation. Sensors (Basel) 18, 1545 (2018). Yoshida, T. et al. Motivation for rehabilitation in patients with subacute stroke: A qualitative study. Front. Rehabil. Sci. 2, 664758 (2021). Kristensen, H. K., Persson, D., Nygren, C., Boll, M. & Matzen, P. Evaluation of evidence within occupational therapy in stroke rehabilitation. Scand. J. Occup. Ther. 18, 11–25 (2011). Steultjens, E. M. J. et al. Occupational therapy for stroke patients: A systematic review. Stroke 34, 676–687 (2003). Djurić-Jovičić, M. et al. Quantification of finger-tapping angle based on wearable sensors. Sensors (Basel) 17, 203 (2017). Yin, C., Liu, Q., Meng, W. & Ai, Q. Quantitative evaluation of hand functions using a wearable glove with multiple sensors, 2021 IEEE International Conference on RCAR. 1093–1098 (2021). Kim, S., Lee, J. & Bae, J. Analysis of finger muscular forces using a wearable hand exoskeleton system. J. Bionic Eng. 14, 680–691 (2017). Zimmerman, T. G., Lanier, J., Blanchard, C., Bryson, S. & Harvill, Y. A hand gesture interface device. ACM Sigchi Bull. 18, 189–192 (1986). Zheng, Y. et al. Development and evaluation of a sensor glove for hand function assessment and preliminary attempts at assessing hand coordination. Measurement 93, 1–12 (2016). Houston, A., Walters, V., Corbett, T. & Coppack, R. Evaluation of a multi-sensor Leap Motion setup for biomechanical motion capture of the hand. J. Biomech. 127, 110713 (2021). Kawaguchi, S., et al . Accuracy evaluation of hand motion measurement using 3D range image sensor, 2017 Eleventh International Conference on Sensing Technology. (2017). Kontson, K. L. et al. Assessing kinematic variability during performance of Jebsen-Taylor Hand Function Test. J. Hand Ther. 33, 34–44 (2020). Kapuscinski, T. & Wysocki, M. Hand gesture recognition for man-machine interaction in Proceedings of the RoMoCo 2001. 91–96 (2001). Yu, C., Wang, X., Huang, H., Shen, J. & Wu, K. Vision-based hand gesture recognition using combinational features. Proceedings of the IIHMSP 2010. 543–546 (2010). Malima, A., Özgür, E. & Çetin, M. A fast algorithm for vision-based hand gesture recognition for robot control. 14th Signal Processing and Communications Applications Conference (IEEE Publications, 2006). Koh, E., Won, J. & Bae, C. On-premise skin color modeling method for vision-based hand tracking. 13th IEEE International Symposium on Consumer Electronics. 908–909 (2009). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-4464229","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":308743593,"identity":"8f0990e3-ff72-4c3d-91c2-bba5c1a8508f","order_by":0,"name":"Na-Yun Seo","email":"","orcid":"","institution":"Chonnam National University","correspondingAuthor":false,"prefix":"","firstName":"Na-Yun","middleName":"","lastName":"Seo","suffix":""},{"id":308743594,"identity":"cba38742-92a6-43f2-80f3-abb7d43e797f","order_by":1,"name":"Joo-Hyun Lee","email":"","orcid":"","institution":"Baekseok University","correspondingAuthor":false,"prefix":"","firstName":"Joo-Hyun","middleName":"","lastName":"Lee","suffix":""},{"id":308743595,"identity":"e41648d3-30f4-4224-bc6a-0842043fbd8b","order_by":2,"name":"Young-Jin Jung","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAoElEQVRIiWNgGAWjYFCCwwcMIAw2orUcSyBZCw9UB9FaDA6e+VBc2MYgz9/AlvaBOC0Hzm4wntnGYDjjANvhGcRr4W1jYNzAwN5MpMMOnHkA0mJPkhYGkJbEDQxsh4nTInngmIExzzmJ5BmH2ZKJ08J34/AzY54yG9v+9jZj4rQo3DjABowYCQYGZuI0MDDI9zcwPyBW8SgYBaNgFIxQAADFiyyI28+euwAAAABJRU5ErkJggg==","orcid":"","institution":"Chonnam National University","correspondingAuthor":true,"prefix":"","firstName":"Young-Jin","middleName":"","lastName":"Jung","suffix":""}],"badges":[],"createdAt":"2024-05-23 04:55:00","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4464229/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4464229/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":58195081,"identity":"d734725e-c54a-4a23-91cf-514092c540c7","added_by":"auto","created_at":"2024-06-12 09:16:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":140436,"visible":true,"origin":"","legend":"\u003cp\u003eExperiment set-up\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/24aa06286b1abf8d11178f38.png"},{"id":58195723,"identity":"f27fd34d-2992-4948-995c-cd7b73d68245","added_by":"auto","created_at":"2024-06-12 09:24:20","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":129759,"visible":true,"origin":"","legend":"\u003cp\u003eHand motion evaluation. (a) Experimental procedure with home-made software. (b) landmark position and labels (which were employed to analyze hand motion quantitatively) are depicted.\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/756e28c32fd1654d0a8d886d.png"},{"id":58195083,"identity":"ee6121de-4efa-4042-972d-681e22f33170","added_by":"auto","created_at":"2024-06-12 09:16:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":111764,"visible":true,"origin":"","legend":"\u003cp\u003eDistance graph. (a) Distance graph of total task. (b) Distance graph with task division. The pillars were used for task division. The line arrow is the BCD section, and the dotted line arrow is the RD section. The small rectangle is the S section. The combination of the line pillar and the dotted pillar is the Pa matrix. The line pillars are the Pb matrix. The X-axis is the number of video frames, and the Y-axis is the distance data.\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/42930d218f717101d57a5fa9.png"},{"id":58195079,"identity":"dd6bacbf-a0bb-4283-8a9c-d92d3bf3e20d","added_by":"auto","created_at":"2024-06-12 09:16:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":62106,"visible":true,"origin":"","legend":"\u003cp\u003eCross product angle of the dorsum of the hand. As the task was done five times, five signals pointing upwards were obtained. When the dorsum of the hand is pointed upwards, the angle is 0. Values closer to 90 ° indicate that the wrist is rotated to the left.\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/c789c89570eda1f527d65195.png"},{"id":58195722,"identity":"abe38a29-4328-48b1-a7dc-039f28cf71e1","added_by":"auto","created_at":"2024-06-12 09:24:20","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":135678,"visible":true,"origin":"","legend":"\u003cp\u003eTrajectories representing hand movement. The blue trajectory is the “transferring bean section.” The green trajectory is the section where beans are put into the cup. The red trajectory is the “return section.” The 5th trajectory’s form is different from others. In the case of the 5th trajectory, the blue is the distance from the 1st bean to the cup. The red is the distance from the cup to the first hand position. The blue trajectory is shorter than that of the red because the blue distance is shorter than the red distance.\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/522465db4294d235b0691259.png"},{"id":71277916,"identity":"882d1caf-35ef-489f-87a6-c1090dc3d80c","added_by":"auto","created_at":"2024-12-12 22:16:31","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":952241,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4464229/v1/03489aed-0d69-4b5f-a40e-adc362b3f9bf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"New Possibilities of A Quantitative Assessment Method for the Jebsen-Taylor Hand Function Test: A Preliminary Study","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eThe Jebsen\u0026ndash;Taylor Hand Function test (JTHFT) was developed in 1969 to offer an objective and standardized assessment of activities of daily living [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The JTHFT is generally used in occupational therapy for patients with various conditions associated with brain injury and is mainly used to evaluate the function of patients with hemiplegia after stroke [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The JTHFT only uses time as a parameter. In other words, the JTHFT focuses on time rather than task performance. Seven tasks are included in the JTHFT: writing a sentence, card turning, lifting small objects, simulated feeding, stacking checkers, picking up light cans, and picking up heavy cans [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOccupational therapy can help patients regain independence in activities of daily living [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Among several assessments, the JTHFT is mainly used in hospitals to evaluate hand function, and its reliability has been proven in several studies [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. In 2015, Mak et al. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] compared the difference between patients with Parkinson\u0026rsquo;s disease and healthy individuals using the JTHFT. Thus, the JTHFT was judged to be a reliable tool that can evaluate the hand function of patients with Parkinson\u0026rsquo;s disease. In 2020, Aynaci et al. [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] reported that objective measures, such as the JTHFT, could help clinicians obtain medical evidence from patients. In 2018, Lim et al. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] measured the reliability of the JTHFT using standard translated instructions and spontaneously translated instructions, reporting its usefulness. In 2021, Lin et al. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] measured the reliability and responsiveness of the JTHFT and BBT in children with cerebral palsy. Both tests showed high reliability when performed within a short time. In 2021, Sığırtma\u0026ccedil; et al. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] proved that JTHFT is a reliable and valid tool and reported that clinicians and researchers could use it to evaluate patients' hand function.\u003c/p\u003e \u003cp\u003eJTHFT can evaluate improved hand function after occupational therapy; however, occupational therapists have identified the following limitations. The results of the JTHFT are only indicated as the time for assessing the agility of the hand. However, this variable cannot determine complex upper-extremity movement patterns such as smoothness, velocity, and coordination during performance assessment. In addition, the change (improvement) in the effects of occupational therapy is too slow to be judged with the naked eye. This can limit a patient's motivation for treatment. In occupational therapy, the patient's passion for participating in voluntary treatment plays an important role. A study revealed a significant difference in functional recovery among patients who had an active willingness to undergo treatment compared to patients who underwent passive treatment. For example, Yoshida et al. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] revealed that patients are greatly motivated when they recognize a change in their body state during the treatment process. Occupational therapy is an essential step for patients, which can relieve many of the inconveniences they experience in daily life and improve their quality of life [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. To motivate patients and restore their ability to perform tasks, occupational therapists have included various compensation strategies for patients during the treatment process [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Occupational therapists must perform several tasks. Occupational therapy after a stroke is time-consuming for occupational therapists and can incur high medical costs or work burdens. Professionalism is the main reason for this increase in medical costs. Particularly, high professionalism is required for therapists to visually judge the degree of the effect of occupational therapy on the patient. To obtain high expertise as occupational therapists, years of patient treatment experience and intensive education are required. This process incurs high costs for training therapists, leading to an increase in medical costs. Consequently, the cost of medical services borne by the patients increases. The high cost of medical services reduces the number of times patients visit the hospital, which reduces the effectiveness of occupational therapy for patients and can decrease their quality of life. In addition to the treatment cost, there is an occupational therapist\u0026rsquo;s work burden, which reduces the quality of life. A high workload in the treatment process causes fatigue in therapists, which can lower the quality of medical services [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. This can negatively affect patients and decrease the treatment effects.\u003c/p\u003e \u003cp\u003eTo overcome the abovementioned limitations and workload of therapists, various techniques for evaluating hand function are being studied. Yin et al. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] automatically evaluated the hand functions using wearable gloves with multiple sensors. Three inertial measurement unit (IMU) sensors and one thin pressure sensor were used to capture the hand movement, and the pressure sensor measured the fingertip pressure. Using this method allowed for the automatic evaluation of the range of motion, flexibility, and scores. Kim et al. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] developed a wearable hand exoskeleton system to analyze the finger muscle forces. The system calculated the load on the fingertips using cable-driven actuators. Zimmerman et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] measured finger flexion using a glove equipped with an analog flex sensor and measured the position and orientation of the hand using ultrasonic and magnetic flux sensors. Zheng et al. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] used sensor gloves to quantitatively measure hand movements. Two types of sensors were used to measure the angle of the finger joint and force exerted by the finger on the object, and their validity and reliability were evaluated. Lin et al. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] developed data gloves for the rehabilitation and assessment of hand function. An IMU sensor and a sensor-fusion algorithm were used in the glove. Its modular design enables more data collection with high scalability. The leaf motion controller (LMC) by Houston et al. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] provided a low-cost method with markerless hand tracking. Several LMCs were used on a single computer to measure hand motions in multiple directions. The Kalman filter and Kabsch algorithm were used, and their reliability and validity were evaluated by comparing the CMC and MJA. Kawaguchi et al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] measured hand movements using an inexpensive and easy-to-use 3D range image sensor. To evaluate the actual measurement accuracy, the error was calculated using measurements from motion capture. The Vicon motion-capture system was used by Kontson et al. [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. The average range of motion and maximum angle of the JTHFT task, performed by attaching a marker to the subjects' bodies, was calculated, and upper-body joint data were collected. Kapuscinski et al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] constructed a gesture-based interface and used skin-colored regions to estimate the hand position. After handshape recognition, gestures were recognized using the hidden Markov model. Yu et al. [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] developed a system for extracting hand-gesture features based on a multilayered perceptron. A total of 3,500 images were used as the dataset, and a system that can recognize hand motion through the Hu-invariant moment and Fourier descriptor was proposed. Malima et al. [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] proposed an algorithm for recognizing motions in hand images. This system comprises three steps: hand region segmentation, finger segmentation, and gesture classification. Koh et al. [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] proposed a vision-based method that can extract the hands without being limited by lighting conditions. To effectively recognize hands in the system, a method for modeling skin color after distinguishing hands from objects was used. These technologies exhibit good performance in evaluating hand function; however, they are insufficient to be used as quantitative evaluation tools instead of the JTHFT by occupational therapists. Although the vision-based method provides accurate positional information with an error of \u0026lt;\u0026thinsp;1 mm, it requires a complex computational process after motion data acquisition. If a quantitative evaluation tool is developed, it can reduce the burden on the therapist and the medical service cost for increasing the quality of life of patients.\u003c/p\u003e \u003cp\u003eTherefore, we examined the possibility of a new quantitative evaluation technique to overcome the aforementioned limitations of JTHFT analysis. In this study, video data were recorded using a webcam. As the motion was recorded using a webcam, the subject could perform the task without interference. We developed custom-made SW software to analyze the video data. SW recognizes the human hand and restores the finger joint from a 2D image to 3D data. Based on the joint position information, various features, such as distance, velocity, time, angle, and trajectory, were extracted. This analysis method uses videos. Therefore, video-based evaluations were possible. Consequently, the evaluation data were preserved. This allows occupational therapists to make evidence-based diagnoses. In addition, homemade SW derives quantitative and objective evaluations. Therefore, patients can judge the extent to which their function has improved. These advantages can reduce the workload of the therapist and motivate patients for treatment. In this paper, \u0026ldquo;Simulated Feeding,\u0026rdquo; which is one of the JTHFT tasks, was analyzed. \u0026ldquo;Simulated Feeding\u0026rdquo; is a task that is most closely related to daily life. In this study, we attempted to confirm the possibility of a quantitative evaluation applicable to occupational therapy assessment.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e This study was approved by the ethics committee of the Institutional Review Board of Baekseok University (BUIRB-202307-HR-020). The study was conducted according to the guidelines of the Declaration of Helsinki, and written informed consent was obtained from the subject before participation.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eExperiment set-up\u003c/h2\u003e \u003cp\u003eA laptop (Legion Y720, Lenovo, China) and a webcam (C920 PRO HD webcam, Logitech, Switzerland) were used to record video images. MATLAB (MATLAB R2022a, MathWorks, TX, USA) was used for program analysis. The recorded video had a resolution of 1280 \u0026times; 720 pixels, and the frame rate was 30 frames per second. When the focus was set to auto, the screen often zoomed in and blurred. Therefore, the focus was set to 1% rather than auto. The homemade application was designed to estimate the quantitative index automatically measured for analysis. To save a video, the name was set using only English and underbar. In this experiment, a webcam was installed on a tripod at a height of 90 cm. The webcam was tilted at a 60-degree. The lens was pointed downwards to confirm the change in distance. The diameter of the beans used in the experiment was 1.6 cm, and the spacing between the beans was 5 cm. Five beans were set side-by-side in a straight line. The beans were set approximately 30 cm away from the subject, and the cup was set approximately 10 cm away from the 1st bean. The right hand was placed approximately 10 cm from the 5th bean. The total distance from the cup to the hand was approximately 60 cm. The hand holding the spoon was 10 cm in diameter (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Data from the same subject performing the task a total of 30 times were used for the analysis. The average video running time was 11.4 s, and the average video size was 26,435 KB. A CSV file containing joint coordinates was used for the analysis. The CSV file contained data extracted from the AI-Motion app using the MediaPipe core. The CSV file contained the values of the X-, Y-, and Z-coordinates corresponding to each joint. There were 21 landmarks for the hand. Therefore, in this file, the number of columns was 63, and the number of rows was the number of video frames. Twenty-one landmarks on the hand were used.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eExperiment task (Task selection)\u003c/h2\u003e \u003cp\u003eSeven tasks are included in JTHFT. Simulated Feeding is the fourth task in the JTHFT. In this study, a Simulated Feeding task that is most closely related to daily life was selected and analyzed. Simulated Feeding involves transferring beans to a cup with a spoon. The experimental method is as follows: (For easy explanation, the bean number on the far left is set to 1, and the bean number on the far right is set to 5)\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe subject grabs the spoon and places his hand 10 cm from the fifth bean.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWhen the video recording starts, the subject waits for about 2 s without moving.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAfter 2 s, the subject places the fifth bean in the spoon and transfers it to a cup. Subsequently, the 4th, 3rd, 2nd, and 1st beans are transferred to the cup.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAfter moving all five beans, the subject returns the spoon to the initial position as the starting point.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAfter returning to the initial position, the subject waits for about 2 s until the video finishes recording.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eDuring this task, the patient\u0026rsquo;s fingers, dorsum of the hand, and wrist were recorded to estimate the 3D joint position from the 2D webcam image.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eEvaluation procedure\u003c/h2\u003e \u003cp\u003eThis study used homemade software to quantitatively evaluate the Simulated Feeding, which was previously qualitatively assessed. This software was developed for the convenience of occupational therapists and can be used without engineering knowledge. The joint points of the hand were extracted using the core of the Mediapipe in the software. Factors that could be analyzed through a Simulated Feeding task, such as the distance, velocity, time, angle, and trajectory, were calculated using this software. The SW processing is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The motion was recorded using a webcam connected to a laptop. The joint points were used as inputs for the task analysis model to calculate the factors. Finally, the analyzed elements were output as a PDF (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (a)).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eDistance calculation\u003c/h2\u003e \u003cp\u003eTo calculate the distance of the task, we used coordinate 12, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (b). The MIDDLE_FINGER_TIP was used as the reference point. This is because the joint has less movement than the other finger joints while holding a spoon. The Euclidean distance formula was used to calculate the distance between two points. The calculated distances are the relative distance values. Thus, we used the ratio of the relative value with the actual value to calculate the actual distance. In this ratio, the relative value is the distance between the first and last points of the task derived from the analysis. The actual value is the distance between the starting point and the cup. This was measured using a tape measure. The actual distance was calculated using two ratios. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (a) illustrates the calculated distance as a graph. The \u0026ldquo;diff function\u0026rdquo; was used to calculate the acceleration values. Acceleration values were used to distinguish the tasks, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e(a). Subsequently, acceleration was classified using a pulse period function. The pulse period function determines the start and end points of the signal. After applying the \u0026ldquo;diff function\u0026rdquo; and \u0026ldquo;pulseperiod function\u0026rdquo; to Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, the signal outputs as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (b).\u003c/p\u003e \u003cp\u003e \u003cdiv id=\"Equ1\" class=\"Equation\"\u003e \u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$Distance= \\sqrt{{({p}_{1}-{q}_{1})}^{2}+{({p}_{2}-{q}_{2})}^{2}+{({p}_{3}-{q}_{3})}^{2}}$$\u003c/div\u003e \u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eCalculation of distance, velocity, and time for each task\u003c/h2\u003e \u003cp\u003eTask classification was calculated using the orange pillars shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e(b). The number of pillars was 11. Therefore, the pillar had a 2 \u0026times; 11 matrix value (Pa matrix), which comprised the start and endpoints. The remaining data, except for those in the first column, are a 2 \u0026times; 10 matrix (=\u0026thinsp;Pb matrix). It contains the start and end data for the task. In this calculation, Pa and Pb matrices were used. A Pa matrix was used to calculate the starting point of the subject\u0026rsquo;s motion. A Pb matrix was used to calculate the values for each task. Simulated Feeding is a task in which the subject transfers beans and returns five times. The beans were transferred to the cup in the \u0026ldquo;transferring bean section.\u0026rdquo; These sections correspond to the 1st, 3rd, 5th, 7th, and 9th columns of the Pb matrix, respectively. In other words, these sections can form a 2 \u0026times; 5 matrix (Po matrix), which is an odd column in the Pb matrix. Similarly, the \u0026ldquo;return section\u0026rdquo; is the returning section after the transferring bean section. These sections correspond to the 2nd, 4th, 6th, 8th, and 10th columns of the Pb matrix, respectively. These sections can form a 2 \u0026times; 5 matrix (Pe matrix), which is an even column in the Pb matrix. Each task involved transfer and return motions. This means that each pillar has odd and even values. With this value, the transferring bean section and return section can be calculated. For example, to calculate the distance of the 1st bean from the cup (1st transferring bean section), the first column of the odd array was used. The distance can be calculated by subtracting Ps(1,1) from Ps(2,1). At this point, the matrix value should be substituted with the distance value calculated in Section 3.1. Therefore, a substituted value was used for subtraction. The calculation of the distance for the \u0026ldquo;return section\u0026rdquo; is the same as above. However, in this study, data from the Pe matrix were used. The time required per task was calculated as the number of frames per 30 iterations. We divided the video by 30 because the frame rate was 30 frames per second. The formula for calculating the velocity was distance/time. The velocity corresponding to each task was calculated using the distance per task/time per task formula.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCross product angle of the dorsum of the hand\u003c/h2\u003e \u003cp\u003eThe cross product was obtained from the dorsum of the hand. The cross-product is a vector pointing upward at 90 \u0026deg; from the dorsum of the hand. There are three areas on the dorsum of the hand: 0-5-9, 0-9-13, and 0-13-17. These areas are created by connecting the landmarks in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (b). The cross product of each area was obtained from these three areas. The average cross-product values for the three areas were calculated. The cross product of the first 10 frames was used to set the reference point. As a result, the cross product was calculated using Eq.\u0026nbsp;\u003cspan refid=\"Equ3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\overrightarrow{a} \\bullet \\overrightarrow{b}= \\left|\\overrightarrow{a}\\right| \\left|\\overrightarrow{b}\\right| cos\\theta$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$cos\\theta = \\frac{\\overrightarrow{a} \\bullet \\overrightarrow{b}}{\\left|\\overrightarrow{a}\\right| \\left|\\overrightarrow{b}\\right|}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\left|\\overrightarrow{a}\\right|= \\sqrt{{V}_{x}^{2}+{V}_{y}^{2}+ {V}_{z}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$$\\overrightarrow{a} \\bullet \\overrightarrow{b}= {a}_{x}{b}_{x}+ {a}_{y}{b}_{y}+ {a}_{z}{b}_{z}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e5\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eTrajectory\u003c/h3\u003e\n\u003cp\u003eThe trajectory represents hand movements for each task. In this calculation, the Pb matrix in 3.2 was also used. Afterward, the \u0026ldquo;scatter3 function\u0026rdquo; and \u0026ldquo;colormap\u0026rdquo; were used. These functions show that the color of the dots changes according to the changes in hand movement.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDistance\u003c/h2\u003e \u003cp\u003eIn Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (a), the part where the graph rises is the \u0026ldquo;transferring bean section.\u0026rdquo; Conversely, the \u0026ldquo;return section\u0026rdquo; is the section where the graph decreases. The graph length of the \u0026ldquo;transferring bean section\u0026rdquo; became shorter with time. This was because the subject's hand gradually moved from the 5th bean to the 1st bean. BCD is the distance of the \u0026ldquo;transferring bean section,\u0026rdquo; implying that BCD is the distance from bean to cup. RD is the distance of the \u0026ldquo;return section,\u0026rdquo; implying that RD is the distance from cup to bean. \u0026ldquo;S\u0026rdquo; is the section where beans are put into a spoon. In Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (b), the line arrow ascending is a section of BCD, and the dotted line arrow descending is the section of RD. The small rectangle is a section of S. The combination of the line and dotted pillars is a Pa matrix. Line pillars represent the Pb matrix. The X-axis represents the number of video frames, and the Y-axis represents distance data.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the results obtained by performing the task 30 times by the same person. The values ​​in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e gradually decrease during the task. Theoretically, the values of the RD (Task 1) and BCD (Task 2) should be identical. This is because the actions occur at the same location. However, the two values ​​ differed. This occurred because the subject performing the task is a human and not a machine. In other words, the subjects moved their hands while trying to put the beans into the spoon. There was no significant difference in the values ​​because the behavioral ranges were short. For example, in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the difference between the RD (Task 1) and BCD (Task 2) was approximately 4 cm. The difference between the RD (Task 2) and BCD (Task 3) was approximately 2 cm. However, in the case of RD, the value for Task 5 increased rapidly. This was because the position of the hand changed. Tasks 1 to 4 involved moving from cup to bean. However, Task 5 involved returning from the cup to the first hand position. Therefore, because task 5 involves returning from the cup to the hand\u0026rsquo;s position where the task first starts, the distance value increases rapidly.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDistance of moving bean to cup and return distance\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eBean to Cup Distance (BCD)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eReturn Distance (RD)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSTD (m)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTables\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e present the velocity and time of moving the bean to the cup and returning, respectively. The velocity is a relative value. Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows that the greater the number of tasks, the shorter the time. This was possibly because the subjects became more accustomed to performing the task, shortening the time.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eVelocity of moving bean to cup and return\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eBean to Cup Velocity (BCV)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eReturn Velocity (RV)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSTD\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTime of moving bean to cup and return\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eBean to Cup Time (BCT)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eReturn Time (RT)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSTD (s)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCross product angle of the dorsum of the hand\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e includes five parts in which the value of the graph increases. This part turns the wrist when the beans are placed in the cup. By specifying the task execution range, the maximum cross-product angle was used to quantify the wrist rotation when performing the task. In Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, the angle was calculated in units of 0.01\u0026deg;. The closer the value of the graph is to 90 \u0026deg;, the more the wrist rotates by approximately 90 \u0026deg;. Through this maximum cross-product angle, the extent to which the subjects rotate their wrist when placing the bean into the cup could be determined. The X-axis represents the number of video frames, and the Y-axis represents the angle data.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCross product angle of the dorsum of the hand\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCross product angle of the dorsum of the hand (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(^\\circ\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e59.27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.47\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e43.66\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eTrajectory\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows the hand movement trajectory for each task. The blue trajectory is the \u0026ldquo;transferring bean section.\u0026rdquo; The green trajectory is the section where the beans were placed in the cup. The red trajectory represents the \u0026ldquo;return section.\u0026rdquo; Five trajectories were extracted because there were five tasks. The shape of the 5th trajectory differed from those of the other four trajectories. For the 5th trajectory, the red trajectory was longer than the blue one. The 5th trajectory is the section where the 1st bean was moved to the cup and returned. The blue trajectory represents the distance from 1st bean to the cup. The red trajectory represents the distance from the cup to the first hand position. The first hand position was next to the 5th bean. The total length of the trajectory is listed in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. Thus, we could calculate the total trajectory length. The values are relative. Based on the calculated value, we could determine whether the subject performed an unnecessary motion. The X-axis represents the number of video frames, and the Y-axis represents the angle data.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRelative length of the trajectory\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRelative length of the trajectory\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.53\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.32\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.46\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eTime\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e lists these two types of times. The first is the time required to place beans in the cup. The second was the time required to place beans in the spoon (BST). The average BCT was approximately 0.174 s. The average BST was approximately 0.438 s. This indicates that the subjects spent more time putting the beans in the spoon than putting it in the cup.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTime of putting bean into the cup and into the spoon\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003ePut Bean into Cup Time (PBCT)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eBean to Spoon Time (BST)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSTD (s)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e, the total time for the task is the time used only for the task (the waiting time of approximately 2 s at the start and end of the task in the experimental method was excluded). In this study, the derived time is 6.82 s. In the case of normal adults (20\u0026ndash;59 s), the time average value is approximately 7.1\u0026thinsp;\u0026plusmn;\u0026thinsp;1.4 s. This value is the standardized time of Simulated Feeding. The time derived in this study includes the normal value range.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTotal task time and normal time\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean (s)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD (s)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNormal (s)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal task time\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.1\u0026thinsp;\u0026plusmn;\u0026thinsp;1.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThe method proposed in this paper quantitatively calculates the elements of the Simulated Feeding of JTHFT using a webcam. The conventional Simulated Feeding was judged only by time. During occupational therapy, the shape of the subject\u0026rsquo;s motion or the angle of hand tilting is impossible to calculate. Moreover, quantitative analysis is not possible. Therefore, it is difficult for patients to confirm that their body function has improved. These limitations make it difficult to motivate patients and can place a burden on occupational therapists. To overcome these limitations, several techniques have been studied to quantitatively evaluate hand function. These techniques can be used to quantitatively evaluate hand function; however, they have certain limitations. For the wearable device, a glove is worn on the hand. This has the disadvantage that external stimuli exist during the performance of the JTFHT. Therefore, patients may not be able to use their full strength during the task. Although low-cost, easy wearability, and high-reliability wearable devices are available, it is difficult for patients with hand discomfort to wear them. In the case of the marker-based method, the marker is attached to the body, and it is inconvenient for patients to perform these tasks. In addition, the markers are expensive. Among the sensors, some have a low cost, such as the LMC. However, this sensor has a narrow field of view and is highly dependent on the sensor position. Basically, this sensor has space constraints. Finally, in the case of most vision-based methods, the results vary depending on lighting because the hand is recognized based on the skin color. To overcome these limitations, we quantitatively analyzed the JTHFT using a homemade SW. As a webcam was used, it was not necessary to attach it to the patient's body. Moreover, they are inexpensive and have no space constraints. In addition, because the SW uses the coordinates of the hand, it is less affected by lighting than conventional vision methods. Using the homemade SW, we quantitatively calculated the elements of JTHFT: distance, velocity, angle, and time. In addition, this SW can schematize the hand's trajectory and the cross-product of the dorsum of the hand, which is otherwise impossible to measure. However, because this study used a single webcam, several issues must be addressed. Recording from top to bottom makes it easy to calculate the distance. However, it is impossible to calculate depth information. In addition, it is difficult to recognize these coordinates. Coordinates were sometimes lost when all fingers were not visible on the webcam screen. However, this issue can be resolved by changing from a single to multiple webcams. If these limitations are resolved, this method can be widely used in occupational therapy, where quantitative judgment is difficult, and in other fields of rehabilitation.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn this paper, the homemade SW quantitatively analyzed Simulated Feeding of JTHFT. The distance, velocity, and time the beans moved were calculated using this SW. In addition, we could calculate the angles of the fingers and the cross-product of the dorsum of the hand, which changed while performing the task. Using a schematic of the trajectory, we determined whether there were any unnecessary movements while the subject performed the task. We confirmed that the total task time derived from SW was included in the average Simulated Feeding time. Thus, we can conclude that the developed SW quantitatively analyzed the Simulated Feeding. We believe that this SW can be used worldwide in areas that require quantitative calculations.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003e \u003cb\u003eCompeting interest statement\u003c/b\u003e:\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eYoung-Jin Jung and Joo-Hun Lee conceived the original idea. Na-Yun Seo wrote the main manuscript text and Young-Jin Jung prepared study design and settings. This was also discussed and reviewed with all authors.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThis work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2022R1C1C1010127)\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and analyzed in the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSığırtma\u0026ccedil;, İ. C. \u0026amp; \u0026Ouml;ks\u0026uuml;z, \u0026Ccedil;. Investigation of reliability, validity, and cutoff value of the Jebsen-Taylor Hand Function Test. J. Hand Ther. 34, 396\u0026ndash;403 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMak, M. K. Y., Lau, E. T. L., Tam, V. W. K., Woo, C. W. Y. \u0026amp; Yuen, S. K. Y. Use of Jebsen Taylor Hand Function Test in evaluating the hand dexterity in people with Parkinson\u0026rsquo;s disease. J. Hand Ther. 28, 389\u0026ndash;394; quiz 395 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLadha, N. \u0026amp; Gaidhani, G. Assessment of hand function in poststroke patients using \u0026lsquo;Jebsen Taylor Hand function test\u0026rsquo;. Indian J. Physiother. Occup. 15, 97\u0026ndash;101 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKwakkel, G., \u003cem\u003eet al\u003c/em\u003e. Effects of augmented exercise therapy time after stroke: A meta-analysis. Stroke 35, 2529\u0026ndash;2539 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSears, E. D. \u0026amp; Chung, K. C. Validity and responsiveness of the Jebsen-Taylor hand function test. J. Hand Surg. Am 35, 30\u0026ndash;37 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAynaci, G. \u0026amp; Kaya, B. \u003cem\u003eEvaluation of Jebsen-Taylor Hand Function Test for Use in Nursing Students: Close-Future Outlook\u003c/em\u003e. International Balkan Education and Science Congress. 13, 711\u0026ndash;718 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLim, Y. X. \u0026amp; Chai, S. C. Standardized translated instruction versus spontaneously translated instruction: test-retest and interrater reliability of a hand function test. J. Hand Ther. 33, 553\u0026ndash;561 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, B. S. \u003cem\u003eet al.\u003c/em\u003e Design of an inertial-sensor-based data glove for hand function evaluation. Sensors (Basel) 18, 1545 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYoshida, T. \u003cem\u003eet al.\u003c/em\u003e Motivation for rehabilitation in patients with subacute stroke: A qualitative study. Front. Rehabil. Sci. 2, 664758 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKristensen, H. K., Persson, D., Nygren, C., Boll, M. \u0026amp; Matzen, P. Evaluation of evidence within occupational therapy in stroke rehabilitation. Scand. J. Occup. Ther. 18, 11\u0026ndash;25 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteultjens, E. M. J. \u003cem\u003eet al.\u003c/em\u003e Occupational therapy for stroke patients: A systematic review. Stroke 34, 676\u0026ndash;687 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDjurić-Jovičić, M. \u003cem\u003eet al.\u003c/em\u003e Quantification of finger-tapping angle based on wearable sensors. Sensors (Basel) 17, 203 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin, C., Liu, Q., Meng, W. \u0026amp; Ai, Q. Quantitative evaluation of hand functions using a wearable glove with multiple sensors, 2021 IEEE International Conference on RCAR. 1093\u0026ndash;1098 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, S., Lee, J. \u0026amp; Bae, J. Analysis of finger muscular forces using a wearable hand exoskeleton system. J. Bionic Eng. 14, 680\u0026ndash;691 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZimmerman, T. G., Lanier, J., Blanchard, C., Bryson, S. \u0026amp; Harvill, Y. A hand gesture interface device. ACM Sigchi Bull. 18, 189\u0026ndash;192 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng, Y. \u003cem\u003eet al.\u003c/em\u003e Development and evaluation of a sensor glove for hand function assessment and preliminary attempts at assessing hand coordination. Measurement 93, 1\u0026ndash;12 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHouston, A., Walters, V., Corbett, T. \u0026amp; Coppack, R. Evaluation of a multi-sensor Leap Motion setup for biomechanical motion capture of the hand. J. Biomech. 127, 110713 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKawaguchi, S., \u003cem\u003eet al\u003c/em\u003e. Accuracy evaluation of hand motion measurement using 3D range image sensor, 2017 Eleventh International Conference on Sensing Technology. (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKontson, K. L. \u003cem\u003eet al.\u003c/em\u003e Assessing kinematic variability during performance of Jebsen-Taylor Hand Function Test. J. Hand Ther. 33, 34\u0026ndash;44 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKapuscinski, T. \u0026amp; Wysocki, M. Hand gesture recognition for man-machine interaction in Proceedings of the RoMoCo 2001. 91\u0026ndash;96 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, C., Wang, X., Huang, H., Shen, J. \u0026amp; Wu, K. Vision-based hand gesture recognition using combinational features. Proceedings of the IIHMSP 2010. 543\u0026ndash;546 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalima, A., \u0026Ouml;zg\u0026uuml;r, E. \u0026amp; \u0026Ccedil;etin, M. A fast algorithm for vision-based hand gesture recognition for robot control. 14th Signal Processing and Communications Applications Conference (IEEE Publications, 2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoh, E., Won, J. \u0026amp; Bae, C. On-premise skin color modeling method for vision-based hand tracking. 13th IEEE International Symposium on Consumer Electronics. 908\u0026ndash;909 (2009).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Deep Learning, Computer-aided Rehabilitation, Hand Function, Motion Trajectory, Quantitative Assessment","lastPublishedDoi":"10.21203/rs.3.rs-4464229/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4464229/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Jebsen\u0026ndash;Taylor Hand Function Test (JTHFT) is a standardized and objective assessment tool for evaluating hand dysfunction in various conditions (including stroke and spinal cord injury). The JTHFT has been widely used successfully in clinical settings; however, the instrument score does not reflect mechanical movement information of the upper extremities. Therefore, we developed a software to quantitatively evaluate the JTHFT. The subject\u0026rsquo;s motion was recorded using a webcam and a deep learning-based three-dimensional motion-extraction technique. Several elements were calculated from vectors between two points. Consequently, eight factors were analyzed: distance to move beans, velocity of the hand moving beans, time to move beans, time to put beans in a spoon, time to put beans in a cup, angle of the dorsum of the hand, trajectory of the hand, and total task time. The average task time was 6.82 s (standard deviation, 0.09), which within the normal range for adults. The cross-product angle of the dorsum of the hand was confirmed to be 0.01 units, depending on how tilted the hand was. In summary, we identified new quantitative assessment tools for occupational therapy; the extracted index results allowed for quantitative evaluation of the aspect that was previously impossible to judge.\u003c/p\u003e","manuscriptTitle":"New Possibilities of A Quantitative Assessment Method for the Jebsen-Taylor Hand Function Test: A Preliminary Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-12 09:16:16","doi":"10.21203/rs.3.rs-4464229/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7e7e1dbe-e5ba-49a2-9732-c79fe93a4552","owner":[],"postedDate":"June 12th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":32618780,"name":"Health sciences/Health occupations"},{"id":32618781,"name":"Physical sciences/Engineering"},{"id":32618782,"name":"Physical sciences/Nanoscience and technology"}],"tags":[],"updatedAt":"2024-12-12T22:08:21+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-12 09:16:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4464229","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4464229","identity":"rs-4464229","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}