Comparison of 3D ankle kinematics between minimal inertial measurement units configuration and optical motion-capture system under diverse walking conditions

Comparison of 3D ankle kinematics between minimal inertial measurement units configuration and optical motion-capture system under diverse walking conditions | 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 Comparison of 3D ankle kinematics between minimal inertial measurement units configuration and optical motion-capture system under diverse walking conditions Jongsu Kim, Lingchao Xie, Sanghyun Cho This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7043663/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 7 You are reading this latest preprint version Abstract Accurate assessment of three-dimensional ankle kinematics is essential for advancing biomechanical analyses and supporting evidence-based clinical interventions. While optical motion-capture systems (OMCS) are considered the gold standard, their high cost and limited portability restrict their use outside the laboratory. Inertial measurement units (IMUs) offer more practical alternative; however, concerns remain regarding their validity and reliability in diverse walking conditions. This study evaluated the validity and repeatability of IMU-derived ankle kinematics relative to OMCS across three walking surfaces: level ground, inward wedge, and outward wedge. Ankle movements in the sagittal, frontal, and transverse planes were analyzed using a minimal sensor configuration. High agreement and repeatability were observed in the sagittal and transverse planes, particularly during level- and inward-inclined walking. Conversely, the frontal plane demonstrated limited agreement under wedged conditions, likely because of complex multisegmental foot dynamics and reference frame misalignment from static calibration. Despite these limitations, consistent repeatability across conditions supports the use of IMUs for tracking ankle kinematics outside the laboratory. These findings suggest that IMUs are applicable as practical tools for clinical and biomechanical studies, particularly where streamlined setups are warranted. Health sciences/Anatomy Physical sciences/Engineering Health sciences/Health care Health sciences/Medical research Optical motion-capture system inertial measurement unit ankle kinematics gait analysis diverse walking conditions validity and repeatability Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Three-dimensional (3D) ankle kinematics are crucial in gait analysis and offer valuable insights for clinical rehabilitation, injury prevention, and biomechanical assessment. Accurate measurement of joint motion enables monitoring of functional recovery by identifying injury risk factors and diagnosing gait abnormalities. 1–3 Therefore, a comprehensive assessment of 3D ankle kinematics is essential for both clinical and research applications. Optical motion capture systems (OMCS), considered as the gold standard for this purpose, provide high accuracy. 4 However, their practical application is limited by the need for numerous markers, high costs, complex data processing, and poor applicability outside laboratory settings. Accordingly, the interest in alternative methods that offer flexibility and practicality has been increasing. Inertial measurement units (IMUs) have emerged as promising substitutes to OMCS, which provide advantages such as portability, cost-effectiveness, and ease of use. 5,6 Unlike OMCS, IMUs require only a single sensor per segment to capture 3D joint kinematics, thereby enabling a significantly simplified setup. 7 This not only lowers costs but also facilitates data processing. Despite these advantages, several challenges hinder the extensive adoption of IMUs. Among the key challenges identified in prior studies, methodological limitations are particularly significant. Substantial research efforts have focused on sagittal plane kinematics, despite the relevance of frontal and transverse planes for understanding functional ankle movement. 3,7 Additionally, the effects of diverse walking conditions have often been overlooked, although such conditions are crucial for evaluating the applicability of IMUs beyond controlled environments. 8,9 Furthermore, while a single sensor per segment is theoretically sufficient to measure joint kinematics, many studies have employed multiple sensors to capture several joints simultaneously. 10,11 These limitations collectively reduce the practical utility of IMU-based assessments, particularly in real-world settings. Therefore, a comprehensive evaluation of the IMU-based assessments is necessary to ensure their generalizability and practical relevance. 12–14 Analyzing ankle kinematics across all three anatomical planes enables the detection of subtle movement variations that reflect the ankle’s multidimensional motions. Measuring performance under varied conditions introduces significant environmental perturbations, which offer insights into sensor sensitivity to external factors. Furthermore, focusing on a single joint while simplifying the sensor setup enhances the efficiency and feasibility of IMU applications. These methodological refinements support the broader use of IMUs beyond laboratory environments. This study evaluated the reliability and validity of 3D ankle kinematics measured by IMUs and OMCS under varied walking conditions by using a minimal sensor setup. We hypothesize that IMU-derived measurements would demonstrate strong agreement with those from OMCS across all three anatomical planes. To test this hypothesis, we quantified the statistical relationship between the two measurement systems by using waveform similarity, performed time-series comparisons, and conducted repeatability analyses. Confirmation of this hypothesis would support the adoption of minimal sensor configurations for accurate ankle kinematic assessments. Methods Participants Twelve healthy young adults (6 males, 6 females; age: 19–33 years; height: 1.58–1.83 m; body mass: 50.6–92.8 kg) participated in this study. The inclusion criteria required the absence of gait abnormalities or lower-extremity musculoskeletal disorders within the past 6 months. The sample size was determined through a power analysis to ensure sufficient statistical robustness for statistical parametric mapping (SPM). 15 Ethical approval was obtained from the Yonsei University Institutional Review Board (IRB No. 1041849-202405-BM-109-03), and all participants provided written informed consent before participation. Protocol A Vicon motion capture system (Vicon Motion Systems Ltd., Oxford, UK) equipped with eight cameras was used as OMCS. Optical markers were placed based on the IOR foot model, which provides anatomically consistent segment alignment for ankle kinematics assessment (Fig. 1). This model is advantageous because of its coordinate system definitions, which are aligned with anatomical bone structures and correspond directly to the anatomical planes of motion. 2 Moreover, its segment-based angle computation is conceptually similar to the IMU algorithm, wherein each segment is represented by a sensor to estimate joint angles. 16 Two IMU sensors (Ultium Motion, Noraxon, AZ, USA) were attached to the shank and dorsal foot following the manufacturer’s guidelines to define segment axes and enable 3D ankle kinematic estimations. 17 This configuration represents the minimal setup required to calculate joint angles. The simplified setup reduces system complexity and user burden, thereby enhancing its practicality for real-world applications. Essentially, it maintains a segment-axis definition comparable to that of the marker-based OMCS. The participants performed static trials to establish OMCS baseline data, with a neutral standing position used to define the reference frames for each marker. 18 For the IMUs, static calibration was performed before each trial to minimize drift and define reference segments essential for precise joint angle estimation. 19 The participants then walked at a self-selected speed across a 3-m walkway under the following three different walking conditions: level floor, inward wedge, and outward wedge (Fig. 2). The bilateral wedges measured 2.25 m in length, 0.35 m in width, and 0.1 m in maximum height, which produced an inclination angle of 16°. Each participant completed 10 walking trials per condition in a randomized order, preceded by three practice trials. To minimize the potential fatigue effects between conditions, a 5-min rest period was provided. Ankle kinematics were simultaneously recorded using both the OMCS and IMUs. The sampling rates for the OMCS and IMUs were set to 200 and 400 Hz, respectively. Data processing Marker data from the OMCS were collected using Vicon Nexus software (Vicon Motion Systems Ltd., Oxford, UK) and processed based on the IOR foot model to estimate the 3D ankle kinematics across the anatomical planes. The captured trajectories were reconstructed and labeled using the IOR foot model. Static trials were performed to establish neutral joint positions, which served as a reference baseline for dynamic measurements. Joint angles were calculated using Python, which applies anatomically aligned rotation matrices to enable precise analysis of 3D ankle kinematics in all three planes of motion. 2,16 ChatGPT-4o (OpenAI, CA, USA) was utilized to verify the Python code used in joint angle calculations. IMU data were processed using myoRESEARCH 3.18 software (MR 3.18, Noraxon, AZ, USA). Signals from the gyroscope, accelerometer, and magnetometer were fused using a quaternion-based algorithm integrated with a Kalman filter to reduce sensor noise and correct drift. This sensor fusion approach enhances the accuracy and reliability of ankle kinematic estimation by leveraging the complementary strength of each sensor to compensate for their individual limitations. The Kalman filter refines the estimates by balancing measurement uncertainty with system dynamics. 17 Gait events were detected using the Z-axis angular velocity data from a gyroscope (Ultium EMG, Noraxon, AZ, USA) mounted on the shank (Fig. 3). Heel strike and toe-off were identified in MR 3.18 based on the characteristic negative peaks in the Z-axis waveform. 20 The gyroscope data were recorded simultaneously with the IMU signals within the same system, which allowed event markers to be directly aligned with the IMU timeline for precise identification of gait events in subsequent analyses. To synchronize the IMU system and OMCS, an accelerometer (Ultium EMG, Noraxon, AZ, USA) and an optical motion capture marker were attached to a rubber hammer. Striking the floor with the hammer generated a distinct peak in both the accelerometer signal and the OMCS marker trajectory. These peaks were used to align timestamps and synchronize the two measurement systems (Fig. 4). After synchronization, data from both systems were time-normalized to 100 frames per gait cycle to facilitate direct comparisons and account for temporal variations in gait. Statistical analysis The consistency between the IMU measurements and OMCS was evaluated using the coefficient of multiple correlation (CMC) across all three anatomical planes and walking conditions. CMC quantifies waveform similarity throughout the gait cycle, thereby offering a comprehensive assessment of kinematic patterns, as opposed to relying on discrete data points. 21,22 The values range from 0.00 to 1.00, with higher CMC scores indicating stronger agreement between systems. This method is particularly well-suited for assessing ankle kinematics under varying walking conditions as it captures the natural variability inherent in human gaits. Paired t-tests using SPM were performed using Python to evaluate the validity of the IMU-derived ankle kinematics via comparison with the OMCS measurements across all three planes. SPM was selected in this study owing to its capacity to identify specific intervals within the gait cycle that exhibit statistically significant differences, which enabled detailed comparisons of continuous kinematic data. In contrast to traditional discrete point analysis, SPM preserves the temporal structure of biomechanical signal and effectively controls Type I error across the entire time series. 23 This makes it particularly well-suited for gait analysis, wherein biomechanical events occur continuously throughout the movement cycle. The repeatability of the IMU measurements was assessed using the intraclass correlation coefficient (ICC), specifically the ICC (2,1) model, which was calculated across repeated trials for each walking condition. ICC was selected for its ability to quantify within-subject measurement consistency, thereby offering insights into the reliability of IMU-derived kinematic data. 24–26 ICC values range from 0.00 to 1.00, with higher values indicating elevated reliability; values above 0.75 are generally considered to reflect excellent reliability. Results Correlation of IMU and OMCS measurements The CMC was calculated to assess waveform similarities between the IMU and OMCS across the three anatomical planes. Correlation strength varied with walking surface and plane of motion. In the sagittal plane, the IMU measurements demonstrated a strong agreement with the OMCS across all conditions. The highest correlation was observed during level walking (CMC = 0.946), indicating significant agreement between the systems. Under the inward wedge condition, correlation remained high (CMC = 0.895), suggesting sagittal plane kinematic were consistently captured despite altered walking surfaces. For the outward wedge condition, the correlation was slightly lower (CMC = 0.797) but reflected acceptable agreement between the two measurement systems. In the frontal plane, level walking demonstrated a relatively low correlation (CMC = 0.273), which indicated limited agreement between the two systems. This correlation decreased further under the inward wedge condition (CMC = 0.154) and highlighted substantial discrepancies in the measurements. Conversely, outward wedge walking showed an improved correlation (CMC = 0.599), which reflected moderate agreement. These findings suggest that IMU-derived measurements in the frontal plane are less consistent with OMCS compared to those in the sagittal plane, with particularly poor concordance observed during inward wedge walking. In the transverse plane, level walking demonstrated a strong correlation (CMC = 0.822), indicating a high agreement between the measurement systems. This strong relationship was maintained under inward wedge condition (CMC = 0.818), thereby suggesting effective transverse plane kinematic capture despite altered surface geometry. However, the correlation decreased significantly during outward wedge walking (CMC = 0.088), which indicated minimal agreement. This significant decrease suggests increased sensitivity of IMU-derived measurements to transverse plane motion changes under outward wedge conditions. Validity of IMU Measurements The SPM analysis revealed significant differences between IMU-derived and OMCS-based ankle kinematics at specific intervals of the gait cycle. In the sagittal plane, significant deviations ( p < 0.05) were observed with the IMUs during level walking, particularly throughout most of the stance and swing phases (14%–94%). In the frontal plane, notable differences were observed during the mid-stance (35%–52%) and early swing phases (56%–65%). As regards the transverse plane, significant variations were noted during the initial contact (0%–5%) and terminal swing (89%–100%). When walking on the inward wedge floor, the IMU recordings showed significant differences ( p < 0.05) in the sagittal plane during terminal stance and throughout the swing phase (44%–86%). In the frontal plane, notable discrepancies were observed during the early stance (0%–45%), mid-swing (56%–67%), and terminal swing (91%–100%). In the transverse plane, significant deviations occurred during terminal stance and early swing phases (52%–72%). During outward floor walking, the IMU data showed significant divergence from the OMCS values across the entire gait cycle in the sagittal plane (0%–100%, p < 0.05). In the frontal plane, discrepancies were observed throughout the stance phase (0%–55%) and during terminal swing phase (73%–100%). Similarly, in the transverse plane, substantial deviations occurred during the stance (0%–53%) and terminal swing phases (75%–100%). Repeatability of IMU Measurements Table 1 Intraclass correlation coefficients (ICC, model 2,1) for IMU-derived ankle kinematics. ICC (95% CI) Level walking Inward tilt wedge floor Outward tilt wedge floor Sagittal plane 0.982 (0.977-0.987) 0.986 (0.987-0.990) 0.953 (0.938-0.965) Frontal plane 0.976 (0.968-0.982) 0.961 (0.949-0.972) 0.933 (0.912-0.954) Transverse plane 0.973 (0.965-0.980) 0.933 (0.912-0.954) 0.958 (0.945-0.969) Mean ICC values and corresponding 95% confidence intervals (CI) are presented for each anatomical plane under the three walking surface conditions. The repeatability of the IMU-derived ankle kinematics was assessed using ICC (2,1) across all walking conditions and anatomical planes (Table 1). During level walking, repeatability was the highest in the sagittal plane (ICC = 0.982), followed by the transverse plane (ICC = 0.976), whereas the frontal plane exhibited slightly lower but high reliability (ICC = 0.973). Under inward wedge floor conditions, repeatability remained high in the sagittal (ICC = 0.986) and frontal (ICC = 0.961) planes, whereas the transverse plane showed a slight decrease (ICC = 0.954), thereby suggesting increased variability in rotational movements. Similarly, during outward wedge floor walking, repeatability remained relatively high in the sagittal (ICC = 0.953) and frontal (ICC = 0.933) planes, whereas the transverse plane (ICC = 0.958) demonstrated comparable reliability. Discussion This study evaluated the agreement and repeatability of IMU-based 3D ankle kinematics relative to the gold standard OMCS across various walking conditions: level ground, inward wedge floor, and outward wedge floor. By examining all three anatomical planes, we assessed waveform similarity using CMC, continuous differences with SPM, and repeatability using the ICC. These analyses were conducted to determine whether a minimal IMU sensor configuration could provide clinically valid ankle kinematic data outside the laboratory. Under level-walking conditions, the sagittal and transverse planes showed relatively high similarity between the IMU and OMCS measurements (CMC = 0.946 and CMC = 0.822, respectively), whereas the frontal plane exhibited a notably lower correlation (CMC = 0.273), indicating weak agreement in that plane. Despite the overall waveform similarity, statistically significant differences ( p < 0.05) were observed: in the sagittal plane between 14–94% of the gait cycle, in the frontal plane during mid-stance (35–52%) and early swing (56–65%), and in the transverse plane during initial contact (0–15%) and terminal swing (89–100%). Although these discrepancies were observed, repeatability remained high across all planes during level walking. These findings align with those of previous studies demonstrating that IMUs can reliably capture 3D ankle kinematics during a typical gait. 2 , 3 , 9 , 26 This consistency strengthens the validity of IMU-based kinematic measurements under level-walking conditions and supports their application in clinical settings where optical systems may not be feasible. Furthermore, this study extends prior research by simultaneously evaluating all three anatomical planes under diverse walking conditions, thereby providing a more comprehensive validation framework. When the participants walked on an inward wedge floor, the sagittal plane maintained a relatively strong correlation (CMC = 0.895) between the IMU and OMCS, with significant differences ( p < 0.05) occurring from 44–86% of the gait cycle. This indicates that the IMUs can capture sagittal plane ankle motion with acceptable accuracy under inward-wedge conditions. Conversely, the frontal plane showed a notably lower correlation (CMC = 0.154) and significant deviations during the early stance (0–45%), mid-swing (56–67%), and terminal swing (91–100%). These discrepancies may result from the inward wedge floor promoting increased ankle inversion, a complex, multisegmental motion that can amplify orientation errors when using only two IMUs. The transverse plane correlation (CMC = 0.818) was comparable to that of level walking, with minor inconsistencies (52–72%). Despite these deviations, the repeatability values remained high (ICC > 0.940) across all the planes, indicating that the IMU-derived measurements were consistently reproducible. The reliability under altered walking conditions supports the feasibility of using minimal IMU setups to monitor ankle kinematics outside the laboratory. The most prominent discrepancies between IMU and OMCS measurements were observed under outward wedge-floor conditions. Although the sagittal plane showed moderate correlation (CMC = 0.797), SPM revealed statistically significant differences ( p < 0.05) across all gait cycle (0–100%). The frontal plane also showed moderate agreement (CMC = 0.599), with significant deviations of 0–55% and 73–100%. Notably, the transverse plane correlation dropped substantially (CMC = 0.088), accompanied by significant differences between 0 and 53% and 75–100% of the gait cycle. This suggests that the rotational movements induced by the outward wedge floor may substantially reduce the agreement between the IMU and OMCS measurements. These discrepancies can be explained by the increased biomechanical complexity introduced by the wedge floor, which may compromise the sensor’s capability to maintain a stable segment orientation. Nevertheless, repeatability remained high across all planes (ICC > 0.912), suggesting that the IMU system provided internally consistent measurements. While the validity of IMU-derived kinematics diminished under complex conditions, the system’s high repeatability supports its applicability for tracking within-subject changes over time in biomechanically challenging conditions. The consistent repeatability across diverse walking conditions suggests that IMUs have the potential to serve as robust tools for longitudinal gait monitoring beyond controlled laboratory environments. Despite this potential, a notable observation across all walking conditions was the consistently lower agreement between the IMU and OMCS measurements in the frontal plane than in the sagittal and transverse planes. This discrepancy may be attributed to several biomechanical and methodological factors. Specifically, ankle inversion and eversion involve multisegmental interactions, particularly at the subtalar joint and midfoot, in contrast to simple uniaxial rotation. 16 , 27 , 28 These complex, multiplanar movements become more pronounced under wedge-walking conditions. However, IMUs, which are typically attached to the shank and dorsal foot, assume rigid-body behavior within each segment. This assumption overlooks the complexity of inter-segmental movements, which limits the capacity to capture multisegmental dynamics. 28 , 29 Therefore, these complex movements could be misrepresented within the local coordinate system of the IMU because of differences in segment definitions compared to the OMCS. These limitations were particularly pronounced during walking on an outward wedge floor. In this condition, depression of the lateral foot induces excessive inversion and increases the demand for supination. The complex interplay among the subtalar joint, midfoot, and shank during these motions may lead to kinematic misrepresentation by the IMUs. Moreover, the simplified sensor configuration permits only static calibration, which is typically performed in an initial standing posture, and assumes a neutral anatomical alignment. However, under wedged walking conditions, this assumption can result in a misalignment between the reference frame and dynamic ankle movements. Such a misalignment may lead to systematic offset errors that accumulate throughout the gait cycle. 30 , 31 Consequently, the agreement between the IMU-based measurements and OMCS was significantly reduced under this condition. The interaction between the biomechanical complexity and simplified sensor configuration could lead to the misrepresentation of 3D ankle kinematics by the IMU. The observed limitations in the frontal plane highlight the need for advanced methodological approaches to improve the interpretation of multisegmental foot and ankle movements. Accurately capturing motion at a single joint typically requires placing at least two IMUs on the joint's adjacent segments. Similarly, reliably tracking complex multisegmental ankle movements under various conditions may require a similar multisensor setup. Additionally, the identified discrepancies under the outward wedge floor condition strongly suggest that dynamic calibration procedures or specialized postprocessing methods are essential for valid IMU-based measurements on uneven surfaces. Therefore, future research should develop adaptive calibration strategies and appropriate postprocessing methods to enhance the reliability and validity of IMUs as portable tools for comprehensive gait analysis. In this study, we hypothesized that IMU-based measurements would demonstrate agreement with OMCS data across all three anatomical planes under varied walking conditions. The results largely supported this hypothesis in the sagittal and transverse planes, particularly during level and inward-inclined walking, wherein correlation and repeatability remained high. However, in the frontal plane, particularly under outward wedge conditions, significant discrepancies emerged, indicating partial limitations in IMU validity under complex biomechanical scenarios. Nonetheless, the observed high repeatability supports the feasibility of both between-subject and within-subject comparisons. These results suggest that a minimal IMU setup may act as a practical tool for capturing ankle kinematics outside laboratory settings. Declarations Acknowledgement We would like to express our sincere gratitude to all those who supported and encouraged us throughout the course of this study. We are also deeply thankful to the participants who voluntarily took part in the experiment. In particular, the first author gratefully acknowledges those who guided and inspired the beginning of their research journey. Author contributions KIM contributed to the conceptualization of the study, conducted the experiment, performed data processing and statistical analysis, and drafted the manuscript. XIE assisted with the experiment, data processing, and statistical analysis. CHO supervised the study and contributed to its conceptualization, experimental design, and manuscript writing. Competing interest The authors declare no competing interests. Data availability The datasets generated and/or analysed during the current study are not publicly available due to concerns regarding participant privacy but are available from the corresponding author upon reasonable request. Ethics declarations Ethical approval was obtained from the Yonsei University Institutional Review Board, and all participants provided written informed consent prior to participation. All procedures were conducted in accordance with the relevant guidelines and regulations. References Baker, R. Gait analysis methods in rehabilitation. J. NeuroEngineering Rehabil. 3 , 4 (2006). Leardini, A. et al. Rear-foot, mid-foot and fore-foot motion during the stance phase of gait. Gait Posture 25 , 453–462 (2007). Park, S. & Yoon, S. Validity Evaluation of an Inertial Measurement Unit (IMU) in Gait Analysis Using Statistical Parametric Mapping (SPM). Sensors 21 , 3667 (2021). Windolf, M., Götzen, N. & Morlock, M. Systematic accuracy and precision analysis of video motion capturing systems—exemplified on the Vicon-460 system. J. Biomech. 41 , 2776–2780 (2008). Chen, S., Lach, J., Lo, B. & Yang, G.-Z. Toward Pervasive Gait Analysis With Wearable Sensors: A Systematic Review. IEEE J. Biomed. Health Inform. 20 , 1521–1537 (2016). Poitras, I. et al. Validity and Reliability of Wearable Sensors for Joint Angle Estimation: A Systematic Review. Sensors 19 , 1555 (2019). Bakhshi, S., Mahoor, M. H. & Davidson, B. S. Development of a body joint angle measurement system using IMU sensors. in 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society 6923–6926 (2011). Renggli, D. et al. Wearable Inertial Measurement Units for Assessing Gait in Real-World Environments. Front. Physiol. 11 , 90 (2020). Piche, E. et al. Validity and repeatability of a new inertial measurement unit system for gait analysis on kinematic parameters: Comparison with an optoelectronic system. Measurement 198 , 111442 (2022). Niswander, W., Wang, W. & Kontson, K. Optimization of IMU Sensor Placement for the Measurement of Lower Limb Joint Kinematics. Sensors 20 , 5993 (2020). Tadano, S., Takeda, R. & Miyagawa, H. Three Dimensional Gait Analysis Using Wearable Acceleration and Gyro Sensors Based on Quaternion Calculations. Sensors 13 , 9321–9343 (2013). Teufl, W., Miezal, M., Taetz, B., Fröhlich, M. & Bleser, G. Validity of inertial sensor based 3D joint kinematics of static and dynamic sport and physiotherapy specific movements. PLOS ONE 14 , e0213064 (2019). Lebleu, J. et al. Lower Limb Kinematics Using Inertial Sensors during Locomotion: Accuracy and Reproducibility of Joint Angle Calculations with Different Sensor-to-Segment Calibrations. Sensors 20 , 715 (2020). Debertin, D., Wargel, A. & Mohr, M. Reliability of Xsens IMU-Based Lower Extremity Joint Angles during In-Field Running. Sensors 24 , 871 (2024). Pataky, T. C. Power1D: a Python toolbox for numerical power estimates in experiments involving one-dimensional continua. PeerJ Comput. Sci. 3 , e125 (2017). Wu, G. et al. ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human joint motion—part I: ankle, hip, and spine. J. Biomech. 35 , 543–548 (2002). Oberländer, K. D. Inertial Measurement Unit (IMU) Technology. https://www.noraxon.com/download/imu-technology-overview/ (2024). Vicon Motion System Ltd. Vicon Nexus User Guide. https://help.vicon.com/space/Nexus216/11605325/Vicon+Nexus+User+Guide (2021). Noraxon. Ultium Motion User Manual. https://www.noraxon.com/download/ultium-motion-user-manual/ (2024). Hori, K. et al. Inertial Measurement Unit-Based Estimation of Foot Trajectory for Clinical Gait Analysis. Front. Physiol. 10 , 1530 (2020). Ferrari, A., Cutti, A. G. & Cappello, A. A new formulation of the coefficient of multiple correlation to assess the similarity of waveforms measured synchronously by different motion analysis protocols. Gait Posture 31 , 540–542 (2010). Røislien, J., Skare, Ø., Opheim, A. & Rennie, L. Evaluating the properties of the coefficient of multiple correlation (CMC) for kinematic gait data. J. Biomech. 45 , 2014–2018 (2012). Pataky, T. C. Generalized n-dimensional biomechanical field analysis using statistical parametric mapping. J. Biomech. 43 , 1976–1982 (2010). Shrout, P. E. & Fleiss, J. L. Intraclass Correlations: Uses in Assessing Rater Reliability. Psychological Bulletin, 86 (2), 420–428 (1979). Al-Amri, M. et al. Inertial Measurement Units for Clinical Movement Analysis: Reliability and Concurrent Validity. Sensors 18 , 719 (2018). Bauer, L., Hamberger, M. A., Böcker, W., Polzer, H. & Baumbach, S. F. Reliability testing of an IMU-based 2-segment foot model for clinical gait analysis. Gait Posture 114 , 112–118 (2024). Nordin, M. & Frankel, H. Basic Biomechanics of the Musculoskeletal System (Lippincott Williams & Wilkins, 2012). Bauer, L., Hamberger, M. A., Böcker, W., Polzer, H. & Baumbach, S. F. Development of an IMU based 2-segment foot model for an applicable medical gait analysis. BMC Musculoskelet. Disord. 25 , 606 (2024). Sekiguchi, Y. et al. Evaluation of the Validity, Reliability, and Kinematic Characteristics of Multi-Segment Foot Models in Motion Capture. Sensors 20 , 4415 (2020). Hu, Q., Liu, L., Mei, F. & Yang, C. Joint Constraints Based Dynamic Calibration of IMU Position on Lower Limbs in IMU-MoCap. Sensors 21 , 7161 (2021). Ekdahl, M., Loewen, A., Erdman, A., Sahin, S. & Ulman, S. Inertial Measurement Unit Sensor-to-Segment Calibration Comparison for Sport-Specific Motion Analysis. Sensors 23 , 7987 (2023). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 11 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Reviewers agreed at journal 12 Jul, 2025 Reviewers agreed at journal 10 Jul, 2025 Reviewers invited by journal 10 Jul, 2025 Editor assigned by journal 10 Jul, 2025 Editor invited by journal 04 Jul, 2025 Submission checks completed at journal 04 Jul, 2025 First submitted to journal 04 Jul, 2025 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-7043663","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":484380866,"identity":"8a6b4ad9-cb40-4677-878f-8526b1a0a1eb","order_by":0,"name":"Jongsu Kim","email":"","orcid":"","institution":"Yonsei University","correspondingAuthor":false,"prefix":"","firstName":"Jongsu","middleName":"","lastName":"Kim","suffix":""},{"id":484380867,"identity":"d94f8fd8-d808-49a3-9d4f-55325c12fd99","order_by":1,"name":"Lingchao Xie","email":"","orcid":"","institution":"Taizhou University","correspondingAuthor":false,"prefix":"","firstName":"Lingchao","middleName":"","lastName":"Xie","suffix":""},{"id":484380868,"identity":"da13f018-e629-4fbb-b75a-c774dfa7dbf5","order_by":2,"name":"Sanghyun Cho","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9klEQVRIiWNgGAWjYBAC9gYQWQDEx8EsBh4QwdiARwvPAZC8AZB15gDJWm4kIETxa2E//vzBDwObxL6bbwwfF/yxkWFgP/yAceYePFp4cgwbewzSEmfezjE2ntmWxsPAk2bAuOEZbi32DDmMzQwGhxM33M4xk+ZtOAz0Sw4D44MDeGzhf/4QouXmGTNpnj//eRj43xDQIpFgCNFygweohe0AD4ME0JYNeLW8MZwJ9IvxzDNpxca8bck8bBLPDA7OwOuw9AcfflTYyPYdP7zxMc8fO3t+/uSHD3vwaEECHAZgig2IidMATDsPiFQ4CkbBKBgFIw0AAFTbUii2iPB5AAAAAElFTkSuQmCC","orcid":"","institution":"Yonsei University","correspondingAuthor":true,"prefix":"","firstName":"Sanghyun","middleName":"","lastName":"Cho","suffix":""}],"badges":[],"createdAt":"2025-07-04 06:38:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7043663/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7043663/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-39161-8","type":"published","date":"2026-02-11T15:58:32+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":86801170,"identity":"83285fd0-bd85-4b15-9c72-0585eea6460d","added_by":"auto","created_at":"2025-07-15 17:04:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":645955,"visible":true,"origin":"","legend":"\u003cp\u003eMarker setup and sensor placement.\u003c/p\u003e\n\u003cp\u003eOptical markers (gray circles) were attached according to the Institute of Rizzoli (IOR) foot model to enable 3D ankle kinematic reconstruction with the optical motion capture system (OMCS). Inertial measurement units (gray rectangles) were affixed to the shank and the dorsal foot to record segment orientations. Additionally, gyroscope (orange rectangle) was mounted on the shank for gait event detection.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/c5110e36f58ac4eb2bbffd8a.png"},{"id":86801171,"identity":"6a0d4a29-f68f-46ae-9701-06793ea6d6e7","added_by":"auto","created_at":"2025-07-15 17:04:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":119967,"visible":true,"origin":"","legend":"\u003cp\u003eWalking surface conditions.\u003c/p\u003e\n\u003cp\u003eLevel floor (a), inward wedge (b), and outward wedge (c). Each platform measured 2.25 m in length, 0.35 m in width, 0.1 m in height. The wedges surfaces were inclined at 16°. Participants completed ten walking trials at a self-selected speed on each surface while kinematic data were recorded.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/42e15fd09449f7a2e96032ba.png"},{"id":86802045,"identity":"57260318-0021-4939-a08a-cc1c0e0cccc1","added_by":"auto","created_at":"2025-07-15 17:12:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":104364,"visible":true,"origin":"","legend":"\u003cp\u003eGait event detection procedure.\u003c/p\u003e\n\u003cp\u003eHeel strike (black circle) and toe off (red circle) events are detected from the Z-axis of angular velocity waveform of the shank mounted gyroscope.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/589d482c107adf94b6da806b.png"},{"id":86802043,"identity":"03a471df-6d24-4263-94d7-da485e6d408c","added_by":"auto","created_at":"2025-07-15 17:12:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":124326,"visible":true,"origin":"","legend":"\u003cp\u003eSynchronization procedure.\u003c/p\u003e\n\u003cp\u003eA hammer strike generates a distinct acceleration peak in the accelerometer signal from the rubber hammer mounted EMG and a simultaneous vertical displacement in the Y-axis trajectory of the collocated optical marker. The corresponding timestamp (red dashed line) is used to synchronize data streams.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/fa5a619ab63a38622e50908c.png"},{"id":86801174,"identity":"21e56a7f-9d0d-4f3b-a137-3962547e5543","added_by":"auto","created_at":"2025-07-15 17:04:48","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":309945,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 4 \u003c/strong\u003eComparison of 3D ankle kinematics between IMU and OMCS across walking conditions.\u003c/p\u003e\n\u003cp\u003eTime series waveform of 3D ankle kinematics measured by the IMUs (red) and the OMCS (black). Each panel displays the CMC, which quantifies waveform similarity between the two systems. Gray bars beneath each plot indicate gait cycle intervals where SPM detected significant differences (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/30e87728fc6368d2c086de5c.png"},{"id":102785851,"identity":"592872fd-f81e-4102-9cae-45a0a0741d55","added_by":"auto","created_at":"2026-02-16 16:10:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1495624,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7043663/v1/70b5e348-4a8f-4f32-b8d8-e86e080695b0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Comparison of 3D ankle kinematics between minimal inertial measurement units configuration and optical motion-capture system under diverse walking conditions","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThree-dimensional (3D) ankle kinematics are crucial in gait analysis and offer valuable insights for clinical rehabilitation, injury prevention, and biomechanical assessment. Accurate measurement of joint motion enables monitoring of functional recovery by identifying injury risk factors and diagnosing gait abnormalities.\u003csup\u003e1\u0026ndash;3\u003c/sup\u003e Therefore, a comprehensive assessment of 3D ankle kinematics is essential for both clinical and research applications. Optical motion capture systems (OMCS), considered as the gold standard for this purpose, provide high accuracy.\u003csup\u003e4\u003c/sup\u003e However, their practical application is limited by the need for numerous markers, high costs, complex data processing, and poor applicability outside laboratory settings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAccordingly, the interest in alternative methods that offer flexibility and practicality has been increasing. Inertial measurement units (IMUs) have emerged as promising substitutes to OMCS, which provide advantages such as portability, cost-effectiveness, and ease of use.\u003csup\u003e5,6\u003c/sup\u003e Unlike OMCS, IMUs require only a single sensor per segment to capture 3D joint kinematics, thereby enabling a significantly simplified setup.\u003csup\u003e7\u003c/sup\u003e This not only lowers costs but also facilitates data processing. Despite these advantages, several challenges hinder the extensive adoption of IMUs.\u003c/p\u003e\n\u003cp\u003eAmong the key challenges identified in prior studies, methodological limitations are particularly significant. Substantial research efforts have focused on sagittal plane kinematics, despite the relevance of frontal and transverse planes for understanding functional ankle movement.\u003csup\u003e3,7\u003c/sup\u003e Additionally, the effects of diverse walking conditions have often been overlooked, although such conditions are crucial for evaluating the applicability of IMUs beyond controlled environments.\u003csup\u003e8,9\u003c/sup\u003e Furthermore, while a single sensor per segment is theoretically sufficient to measure joint kinematics, many studies have employed multiple sensors to capture several joints simultaneously.\u003csup\u003e10,11\u003c/sup\u003e These limitations collectively reduce the practical utility of IMU-based assessments, particularly in real-world settings.\u003c/p\u003e\n\u003cp\u003eTherefore, a comprehensive evaluation of the IMU-based assessments is necessary to ensure their generalizability and practical relevance.\u003csup\u003e12\u0026ndash;14\u003c/sup\u003e Analyzing ankle kinematics across all three anatomical planes enables the detection of subtle movement variations that reflect the ankle\u0026rsquo;s multidimensional motions. Measuring performance under varied conditions introduces significant environmental perturbations, which offer insights into sensor sensitivity to external factors. Furthermore, focusing on a single joint while simplifying the sensor setup enhances\u0026nbsp;the efficiency and feasibility of IMU applications. These methodological refinements support the broader use of IMUs beyond\u0026nbsp;laboratory\u0026nbsp;environments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study evaluated the reliability and validity of 3D ankle kinematics measured by IMUs and OMCS under varied walking conditions by using a minimal sensor setup. We hypothesize that IMU-derived measurements would demonstrate strong agreement with those from OMCS across all three anatomical planes. To test this hypothesis, we quantified the statistical relationship between the two measurement systems by using waveform similarity, performed time-series comparisons, and conducted repeatability analyses. Confirmation of this hypothesis would support the adoption of minimal sensor configurations for accurate ankle kinematic assessments.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eParticipants\u003c/p\u003e\n\u003cp\u003eTwelve healthy young adults (6 males, 6 females; age: 19\u0026ndash;33 years; height: 1.58\u0026ndash;1.83 m; body mass: 50.6\u0026ndash;92.8 kg) participated in this study. The inclusion criteria required the absence of gait abnormalities or lower-extremity musculoskeletal disorders within the past 6 months. The sample size was determined through a power analysis to ensure sufficient statistical robustness for statistical parametric mapping (SPM).\u003csup\u003e15\u003c/sup\u003e Ethical approval was obtained from the Yonsei University Institutional Review Board (IRB No. 1041849-202405-BM-109-03), and all participants provided written informed consent before participation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eProtocol\u003c/p\u003e\n\u003cp\u003eA Vicon motion capture system (Vicon Motion Systems Ltd., Oxford, UK) equipped with eight cameras was used as OMCS. Optical markers were placed based on the IOR foot model, which provides anatomically consistent segment alignment for ankle kinematics assessment (Fig. 1). This model is advantageous because of its coordinate system definitions, which are aligned with anatomical bone structures and correspond directly to the anatomical planes of motion.\u003csup\u003e2\u003c/sup\u003e Moreover, its segment-based angle computation is conceptually similar to the IMU algorithm, wherein each segment is represented by a sensor to estimate joint angles.\u003csup\u003e16\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eTwo IMU sensors (Ultium Motion, Noraxon, AZ, USA) were attached to the shank and dorsal foot following the manufacturer\u0026rsquo;s guidelines to define segment axes and enable 3D ankle kinematic estimations.\u003csup\u003e17\u003c/sup\u003e This configuration represents the minimal setup required to calculate joint angles. The simplified setup reduces system complexity and user burden, thereby enhancing its practicality for real-world applications. Essentially, it maintains a segment-axis definition comparable to that of the marker-based OMCS.\u003c/p\u003e\n\u003cp\u003eThe participants performed static trials to establish OMCS baseline data, with a neutral standing position used to define the reference frames for each marker.\u003csup\u003e18\u003c/sup\u003e For the IMUs, static calibration was performed before each trial to minimize drift and define reference segments essential for precise joint angle estimation.\u003csup\u003e19\u003c/sup\u003e The participants then walked at a self-selected speed across a 3-m walkway under the following three different walking conditions: level floor, inward wedge, and outward wedge (Fig. 2). The bilateral wedges measured 2.25 m in length, 0.35 m in width, and 0.1 m in maximum height, which produced an inclination angle of 16\u0026deg;. Each participant completed 10 walking trials per condition in a randomized order, preceded by three practice trials. To minimize the potential fatigue effects between conditions, a 5-min rest period was provided. Ankle kinematics were simultaneously recorded using both the OMCS and IMUs. The sampling rates for the OMCS and IMUs were set to 200 and 400 Hz, respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData processing\u003c/p\u003e\n\u003cp\u003eMarker data from the OMCS were collected using Vicon Nexus software (Vicon Motion Systems Ltd., Oxford, UK) and processed based on the IOR foot model to estimate the 3D ankle kinematics across the anatomical planes. The captured trajectories were reconstructed and labeled using the IOR foot model. Static trials were performed to establish neutral joint positions, which served as a reference baseline for dynamic measurements. Joint angles were calculated using Python, which applies anatomically aligned rotation matrices to enable precise analysis of 3D ankle kinematics in all three planes of motion.\u003csup\u003e2,16\u0026nbsp;\u003c/sup\u003eChatGPT-4o (OpenAI, CA, USA) was utilized to verify the Python code used in joint angle calculations.\u003c/p\u003e\n\u003cp\u003eIMU data were processed using myoRESEARCH 3.18 software (MR 3.18, Noraxon, AZ, USA). Signals from the gyroscope, accelerometer, and magnetometer were fused using a quaternion-based algorithm integrated with a Kalman filter to reduce sensor noise and correct drift. This sensor fusion approach enhances the accuracy and reliability of ankle kinematic estimation by leveraging the complementary strength of each sensor to compensate for their individual limitations. The Kalman filter refines the estimates by balancing measurement uncertainty with system dynamics.\u003csup\u003e17\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eGait events were detected using the Z-axis angular velocity data from a gyroscope (Ultium EMG, Noraxon, AZ, USA) mounted on the shank (Fig. 3). Heel strike and toe-off were identified in MR 3.18 based on the characteristic negative peaks in the Z-axis waveform.\u003csup\u003e20\u003c/sup\u003e The gyroscope data were recorded simultaneously with the IMU signals within the same system, which allowed event markers to be directly aligned with the IMU timeline for precise identification of gait events in subsequent analyses.\u003c/p\u003e\n\u003cp\u003eTo synchronize the IMU system and OMCS, an accelerometer (Ultium EMG, Noraxon, AZ, USA) and an optical motion capture marker were attached to a rubber hammer. Striking the floor with the hammer generated a distinct peak in both the accelerometer signal and the OMCS marker trajectory. These peaks were used to align timestamps and synchronize the two measurement systems (Fig. 4). After synchronization, data from both systems were time-normalized to 100 frames per gait cycle to facilitate direct comparisons and account for temporal variations in gait.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eStatistical analysis\u003c/p\u003e\n\u003cp\u003eThe consistency between the IMU measurements and OMCS was evaluated using the coefficient of multiple correlation (CMC) across all three anatomical planes and walking conditions. CMC quantifies waveform similarity throughout the gait cycle, thereby offering a comprehensive assessment of kinematic patterns, as opposed to relying on discrete data points.\u003csup\u003e21,22\u003c/sup\u003e The values range from 0.00 to 1.00, with higher CMC scores indicating stronger agreement between systems. This method is particularly well-suited for assessing ankle kinematics under varying walking conditions as it captures the natural variability inherent in human gaits.\u003c/p\u003e\n\u003cp\u003ePaired t-tests using SPM were performed using Python to evaluate the validity of the IMU-derived ankle kinematics via comparison with the OMCS measurements across all three planes. SPM was selected in this study owing to its capacity to identify specific intervals within the gait cycle that exhibit statistically significant differences, which enabled detailed comparisons of continuous kinematic data. In contrast to traditional discrete point analysis, SPM preserves the temporal structure of biomechanical signal and effectively controls Type I error across the entire time series.\u003csup\u003e23\u003c/sup\u003e This makes it particularly well-suited for gait analysis, wherein biomechanical events occur continuously throughout the movement cycle.\u003c/p\u003e\n\u003cp\u003eThe repeatability of the IMU measurements was assessed using the intraclass correlation coefficient (ICC), specifically the ICC (2,1) model, which was calculated across repeated trials for each walking condition. ICC was selected for its ability to quantify within-subject measurement consistency, thereby offering insights into the reliability of IMU-derived kinematic data.\u003csup\u003e24\u0026ndash;26\u003c/sup\u003e ICC values range from 0.00 to 1.00, with higher values indicating elevated reliability; values above 0.75 are generally considered to reflect excellent reliability.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eCorrelation of IMU and OMCS measurements\u003c/p\u003e\n\u003cp\u003eThe CMC was calculated to assess waveform similarities between the IMU and OMCS across the three anatomical planes. Correlation strength varied with walking surface and plane of motion. In the sagittal plane, the IMU measurements demonstrated a strong agreement with the OMCS across all conditions. The highest correlation was observed during level walking (CMC = 0.946), indicating significant agreement between the systems. Under the inward wedge condition, correlation remained high (CMC = 0.895), suggesting sagittal plane kinematic were consistently captured despite altered walking surfaces. For the outward wedge condition, the correlation was slightly lower (CMC = 0.797) but reflected acceptable agreement between the two measurement systems.\u003c/p\u003e\n\u003cp\u003eIn the frontal plane, level walking demonstrated a relatively low correlation (CMC = 0.273), which indicated limited agreement between the two systems. This correlation decreased further under the inward wedge condition (CMC = 0.154) and highlighted substantial discrepancies in the measurements. Conversely, outward wedge walking showed an improved correlation (CMC = 0.599), which reflected moderate agreement. These findings suggest that IMU-derived measurements in the frontal plane are less consistent with OMCS compared to those in the sagittal plane, with particularly poor concordance observed during inward wedge walking.\u003c/p\u003e\n\u003cp\u003eIn the transverse plane, level walking demonstrated a strong correlation (CMC = 0.822), indicating a high agreement between the measurement systems. This strong relationship was maintained under inward wedge condition (CMC = 0.818), thereby suggesting effective transverse plane kinematic capture despite altered surface geometry. However, the correlation decreased significantly during outward wedge walking (CMC = 0.088), which indicated minimal agreement. This significant decrease suggests increased sensitivity of IMU-derived measurements to transverse plane motion changes under outward wedge conditions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eValidity of IMU Measurements\u003c/p\u003e\n\u003cp\u003eThe SPM analysis revealed significant differences between IMU-derived and OMCS-based ankle kinematics at specific intervals of the gait cycle. In the sagittal plane, significant deviations (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u0026lt; 0.05) were observed with the IMUs during level walking, particularly throughout most of the stance and swing phases (14%\u0026ndash;94%). In the frontal plane, notable differences were observed during the mid-stance (35%\u0026ndash;52%) and early swing phases (56%\u0026ndash;65%). As regards the transverse plane, significant variations were noted during the initial contact (0%\u0026ndash;5%) and terminal swing (89%\u0026ndash;100%).\u003c/p\u003e\n\u003cp\u003eWhen walking on the inward wedge floor, the IMU recordings showed significant differences (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) in the sagittal plane during terminal stance and throughout the swing phase (44%\u0026ndash;86%). In the frontal plane, notable discrepancies were observed during the early stance (0%\u0026ndash;45%), mid-swing (56%\u0026ndash;67%), and terminal swing (91%\u0026ndash;100%). In the transverse plane, significant deviations occurred during terminal stance and early swing phases (52%\u0026ndash;72%).\u003c/p\u003e\n\u003cp\u003eDuring outward floor walking, the IMU data showed significant divergence from the OMCS values across the entire gait cycle in the sagittal plane (0%\u0026ndash;100%,\u003cem\u003e\u0026nbsp;p\u003c/em\u003e \u0026lt; 0.05). In the frontal plane, discrepancies were observed throughout the stance phase (0%\u0026ndash;55%) and during terminal swing phase (73%\u0026ndash;100%). Similarly, in the transverse plane, substantial deviations occurred during the stance (0%\u0026ndash;53%) and terminal swing phases (75%\u0026ndash;100%).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRepeatability of IMU Measurements\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1 Intraclass correlation coefficients (ICC, model 2,1) for IMU-derived ankle kinematics.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eICC (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eLevel walking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eInward tilt\u003cbr\u003e\u0026nbsp;wedge floor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eOutward tilt\u003cbr\u003e\u0026nbsp;wedge floor\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eSagittal plane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.982 (0.977-0.987)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.986 (0.987-0.990)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.953 (0.938-0.965)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eFrontal plane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.976 (0.968-0.982)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.961 (0.949-0.972)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.933 (0.912-0.954)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eTransverse plane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.973 (0.965-0.980)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.933 (0.912-0.954)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e0.958 (0.945-0.969)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eMean ICC values and corresponding 95% confidence intervals (CI) are presented for each anatomical plane under the three walking surface conditions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe repeatability of the IMU-derived ankle kinematics was assessed using ICC (2,1) across all walking conditions and anatomical planes (Table 1). During level walking, repeatability was the highest in the sagittal plane (ICC = 0.982), followed by the transverse plane (ICC = 0.976), whereas the frontal plane exhibited slightly lower but high reliability (ICC = 0.973).\u003c/p\u003e\n\u003cp\u003eUnder inward wedge floor conditions, repeatability remained high in the sagittal (ICC = 0.986) and frontal (ICC = 0.961) planes, whereas the transverse plane showed a slight decrease (ICC = 0.954), thereby suggesting increased variability in rotational movements.\u003c/p\u003e\n\u003cp\u003eSimilarly, during outward wedge floor walking, repeatability remained relatively high in the sagittal (ICC = 0.953) and frontal (ICC = 0.933) planes, whereas the transverse plane (ICC = 0.958) demonstrated comparable reliability.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study evaluated the agreement and repeatability of IMU-based 3D ankle kinematics relative to the gold standard OMCS across various walking conditions: level ground, inward wedge floor, and outward wedge floor. By examining all three anatomical planes, we assessed waveform similarity using CMC, continuous differences with SPM, and repeatability using the ICC. These analyses were conducted to determine whether a minimal IMU sensor configuration could provide clinically valid ankle kinematic data outside the laboratory.\u003c/p\u003e\u003cp\u003eUnder level-walking conditions, the sagittal and transverse planes showed relatively high similarity between the IMU and OMCS measurements (CMC\u0026thinsp;=\u0026thinsp;0.946 and CMC\u0026thinsp;=\u0026thinsp;0.822, respectively), whereas the frontal plane exhibited a notably lower correlation (CMC\u0026thinsp;=\u0026thinsp;0.273), indicating weak agreement in that plane. Despite the overall waveform similarity, statistically significant differences (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were observed: in the sagittal plane between 14\u0026ndash;94% of the gait cycle, in the frontal plane during mid-stance (35\u0026ndash;52%) and early swing (56\u0026ndash;65%), and in the transverse plane during initial contact (0\u0026ndash;15%) and terminal swing (89\u0026ndash;100%). Although these discrepancies were observed, repeatability remained high across all planes during level walking. These findings align with those of previous studies demonstrating that IMUs can reliably capture 3D ankle kinematics during a typical gait.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e This consistency strengthens the validity of IMU-based kinematic measurements under level-walking conditions and supports their application in clinical settings where optical systems may not be feasible. Furthermore, this study extends prior research by simultaneously evaluating all three anatomical planes under diverse walking conditions, thereby providing a more comprehensive validation framework.\u003c/p\u003e\u003cp\u003eWhen the participants walked on an inward wedge floor, the sagittal plane maintained a relatively strong correlation (CMC\u0026thinsp;=\u0026thinsp;0.895) between the IMU and OMCS, with significant differences (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) occurring from 44\u0026ndash;86% of the gait cycle. This indicates that the IMUs can capture sagittal plane ankle motion with acceptable accuracy under inward-wedge conditions. Conversely, the frontal plane showed a notably lower correlation (CMC\u0026thinsp;=\u0026thinsp;0.154) and significant deviations during the early stance (0\u0026ndash;45%), mid-swing (56\u0026ndash;67%), and terminal swing (91\u0026ndash;100%). These discrepancies may result from the inward wedge floor promoting increased ankle inversion, a complex, multisegmental motion that can amplify orientation errors when using only two IMUs. The transverse plane correlation (CMC\u0026thinsp;=\u0026thinsp;0.818) was comparable to that of level walking, with minor inconsistencies (52\u0026ndash;72%). Despite these deviations, the repeatability values remained high (ICC\u0026thinsp;\u0026gt;\u0026thinsp;0.940) across all the planes, indicating that the IMU-derived measurements were consistently reproducible. The reliability under altered walking conditions supports the feasibility of using minimal IMU setups to monitor ankle kinematics outside the laboratory.\u003c/p\u003e\u003cp\u003eThe most prominent discrepancies between IMU and OMCS measurements were observed under outward wedge-floor conditions. Although the sagittal plane showed moderate correlation (CMC\u0026thinsp;=\u0026thinsp;0.797), SPM revealed statistically significant differences (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) across all gait cycle (0\u0026ndash;100%). The frontal plane also showed moderate agreement (CMC\u0026thinsp;=\u0026thinsp;0.599), with significant deviations of 0\u0026ndash;55% and 73\u0026ndash;100%. Notably, the transverse plane correlation dropped substantially (CMC\u0026thinsp;=\u0026thinsp;0.088), accompanied by significant differences between 0 and 53% and 75\u0026ndash;100% of the gait cycle. This suggests that the rotational movements induced by the outward wedge floor may substantially reduce the agreement between the IMU and OMCS measurements. These discrepancies can be explained by the increased biomechanical complexity introduced by the wedge floor, which may compromise the sensor\u0026rsquo;s capability to maintain a stable segment orientation. Nevertheless, repeatability remained high across all planes (ICC\u0026thinsp;\u0026gt;\u0026thinsp;0.912), suggesting that the IMU system provided internally consistent measurements. While the validity of IMU-derived kinematics diminished under complex conditions, the system\u0026rsquo;s high repeatability supports its applicability for tracking within-subject changes over time in biomechanically challenging conditions.\u003c/p\u003e\u003cp\u003eThe consistent repeatability across diverse walking conditions suggests that IMUs have the potential to serve as robust tools for longitudinal gait monitoring beyond controlled laboratory environments. Despite this potential, a notable observation across all walking conditions was the consistently lower agreement between the IMU and OMCS measurements in the frontal plane than in the sagittal and transverse planes. This discrepancy may be attributed to several biomechanical and methodological factors. Specifically, ankle inversion and eversion involve multisegmental interactions, particularly at the subtalar joint and midfoot, in contrast to simple uniaxial rotation.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e These complex, multiplanar movements become more pronounced under wedge-walking conditions. However, IMUs, which are typically attached to the shank and dorsal foot, assume rigid-body behavior within each segment. This assumption overlooks the complexity of inter-segmental movements, which limits the capacity to capture multisegmental dynamics.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e Therefore, these complex movements could be misrepresented within the local coordinate system of the IMU because of differences in segment definitions compared to the OMCS.\u003c/p\u003e\u003cp\u003eThese limitations were particularly pronounced during walking on an outward wedge floor. In this condition, depression of the lateral foot induces excessive inversion and increases the demand for supination. The complex interplay among the subtalar joint, midfoot, and shank during these motions may lead to kinematic misrepresentation by the IMUs. Moreover, the simplified sensor configuration permits only static calibration, which is typically performed in an initial standing posture, and assumes a neutral anatomical alignment. However, under wedged walking conditions, this assumption can result in a misalignment between the reference frame and dynamic ankle movements. Such a misalignment may lead to systematic offset errors that accumulate throughout the gait cycle.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e Consequently, the agreement between the IMU-based measurements and OMCS was significantly reduced under this condition. The interaction between the biomechanical complexity and simplified sensor configuration could lead to the misrepresentation of 3D ankle kinematics by the IMU.\u003c/p\u003e\u003cp\u003eThe observed limitations in the frontal plane highlight the need for advanced methodological approaches to improve the interpretation of multisegmental foot and ankle movements. Accurately capturing motion at a single joint typically requires placing at least two IMUs on the joint's adjacent segments. Similarly, reliably tracking complex multisegmental ankle movements under various conditions may require a similar multisensor setup. Additionally, the identified discrepancies under the outward wedge floor condition strongly suggest that dynamic calibration procedures or specialized postprocessing methods are essential for valid IMU-based measurements on uneven surfaces. Therefore, future research should develop adaptive calibration strategies and appropriate postprocessing methods to enhance the reliability and validity of IMUs as portable tools for comprehensive gait analysis.\u003c/p\u003e\u003cp\u003eIn this study, we hypothesized that IMU-based measurements would demonstrate agreement with OMCS data across all three anatomical planes under varied walking conditions. The results largely supported this hypothesis in the sagittal and transverse planes, particularly during level and inward-inclined walking, wherein correlation and repeatability remained high. However, in the frontal plane, particularly under outward wedge conditions, significant discrepancies emerged, indicating partial limitations in IMU validity under complex biomechanical scenarios. Nonetheless, the observed high repeatability supports the feasibility of both between-subject and within-subject comparisons. These results suggest that a minimal IMU setup may act as a practical tool for capturing ankle kinematics outside laboratory settings.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to express our sincere gratitude to all those who supported and encouraged us throughout the course of this study. We are also deeply thankful to the participants who voluntarily took part in the experiment. In particular, the first author gratefully acknowledges those who guided and inspired the beginning of their research journey.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKIM contributed to the conceptualization of the study, conducted the experiment, performed data processing and statistical analysis, and drafted the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eXIE assisted with the experiment, data processing, and statistical analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCHO supervised the study and contributed to its conceptualization, experimental design, and manuscript writing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and/or analysed during the current study are not publicly available due to concerns regarding participant privacy but are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was obtained from the Yonsei University Institutional Review Board, and all participants provided written informed consent prior to participation. All procedures were conducted in accordance with the relevant guidelines and regulations.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBaker, R. Gait analysis methods in rehabilitation. \u003cem\u003eJ. NeuroEngineering Rehabil.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 4 (2006).\u003c/li\u003e\n\u003cli\u003eLeardini, A. \u003cem\u003eet al.\u003c/em\u003e Rear-foot, mid-foot and fore-foot motion during the stance phase of gait. \u003cem\u003eGait Posture\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 453\u0026ndash;462 (2007).\u003c/li\u003e\n\u003cli\u003ePark, S. \u0026amp; Yoon, S. Validity Evaluation of an Inertial Measurement Unit (IMU) in Gait Analysis Using Statistical Parametric Mapping (SPM). \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 3667 (2021).\u003c/li\u003e\n\u003cli\u003eWindolf, M., G\u0026ouml;tzen, N. \u0026amp; Morlock, M. Systematic accuracy and precision analysis of video motion capturing systems\u0026mdash;exemplified on the Vicon-460 system. \u003cem\u003eJ. Biomech.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 2776\u0026ndash;2780 (2008).\u003c/li\u003e\n\u003cli\u003eChen, S., Lach, J., Lo, B. \u0026amp; Yang, G.-Z. Toward Pervasive Gait Analysis With Wearable Sensors: A Systematic Review. \u003cem\u003eIEEE J. Biomed. Health Inform.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1521\u0026ndash;1537 (2016).\u003c/li\u003e\n\u003cli\u003ePoitras, I. \u003cem\u003eet al.\u003c/em\u003e Validity and Reliability of Wearable Sensors for Joint Angle Estimation: A Systematic Review. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 1555 (2019).\u003c/li\u003e\n\u003cli\u003eBakhshi, S., Mahoor, M. H. \u0026amp; Davidson, B. S. Development of a body joint angle measurement system using IMU sensors. in \u003cem\u003e2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society\u003c/em\u003e 6923\u0026ndash;6926 (2011).\u003c/li\u003e\n\u003cli\u003eRenggli, D. \u003cem\u003eet al.\u003c/em\u003e Wearable Inertial Measurement Units for Assessing Gait in Real-World Environments. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 90 (2020).\u003c/li\u003e\n\u003cli\u003ePiche, E. \u003cem\u003eet al.\u003c/em\u003e Validity and repeatability of a new inertial measurement unit system for gait analysis on kinematic parameters: Comparison with an optoelectronic system. \u003cem\u003eMeasurement\u003c/em\u003e \u003cstrong\u003e198\u003c/strong\u003e, 111442 (2022).\u003c/li\u003e\n\u003cli\u003eNiswander, W., Wang, W. \u0026amp; Kontson, K. Optimization of IMU Sensor Placement for the Measurement of Lower Limb Joint Kinematics. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 5993 (2020).\u003c/li\u003e\n\u003cli\u003eTadano, S., Takeda, R. \u0026amp; Miyagawa, H. Three Dimensional Gait Analysis Using Wearable Acceleration and Gyro Sensors Based on Quaternion Calculations. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 9321\u0026ndash;9343 (2013).\u003c/li\u003e\n\u003cli\u003eTeufl, W., Miezal, M., Taetz, B., Fr\u0026ouml;hlich, M. \u0026amp; Bleser, G. Validity of inertial sensor based 3D joint kinematics of static and dynamic sport and physiotherapy specific movements. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e0213064 (2019).\u003c/li\u003e\n\u003cli\u003eLebleu, J. \u003cem\u003eet al.\u003c/em\u003e Lower Limb Kinematics Using Inertial Sensors during Locomotion: Accuracy and Reproducibility of Joint Angle Calculations with Different Sensor-to-Segment Calibrations. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 715 (2020).\u003c/li\u003e\n\u003cli\u003eDebertin, D., Wargel, A. \u0026amp; Mohr, M. Reliability of Xsens IMU-Based Lower Extremity Joint Angles during In-Field Running. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 871 (2024).\u003c/li\u003e\n\u003cli\u003ePataky, T. C. Power1D: a Python toolbox for numerical power estimates in experiments involving one-dimensional continua. \u003cem\u003ePeerJ Comput. Sci.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e125 (2017).\u003c/li\u003e\n\u003cli\u003eWu, G. \u003cem\u003eet al.\u003c/em\u003e ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human joint motion\u0026mdash;part I: ankle, hip, and spine. \u003cem\u003eJ. Biomech.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 543\u0026ndash;548 (2002).\u003c/li\u003e\n\u003cli\u003eOberl\u0026auml;nder, K. D. Inertial Measurement Unit (IMU) Technology. https://www.noraxon.com/download/imu-technology-overview/ (2024).\u003c/li\u003e\n\u003cli\u003eVicon Motion System Ltd. Vicon Nexus User Guide. https://help.vicon.com/space/Nexus216/11605325/Vicon+Nexus+User+Guide (2021).\u003c/li\u003e\n\u003cli\u003eNoraxon. Ultium Motion User Manual. https://www.noraxon.com/download/ultium-motion-user-manual/ (2024).\u003c/li\u003e\n\u003cli\u003eHori, K. \u003cem\u003eet al.\u003c/em\u003e Inertial Measurement Unit-Based Estimation of Foot Trajectory for Clinical Gait Analysis. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1530 (2020).\u003c/li\u003e\n\u003cli\u003eFerrari, A., Cutti, A. G. \u0026amp; Cappello, A. A new formulation of the coefficient of multiple correlation to assess the similarity of waveforms measured synchronously by different motion analysis protocols. \u003cem\u003eGait Posture\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 540\u0026ndash;542 (2010).\u003c/li\u003e\n\u003cli\u003eR\u0026oslash;islien, J., Skare, \u0026Oslash;., Opheim, A. \u0026amp; Rennie, L. Evaluating the properties of the coefficient of multiple correlation (CMC) for kinematic gait data. \u003cem\u003eJ. Biomech.\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 2014\u0026ndash;2018 (2012).\u003c/li\u003e\n\u003cli\u003ePataky, T. C. Generalized n-dimensional biomechanical field analysis using statistical parametric mapping. \u003cem\u003eJ. Biomech.\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 1976\u0026ndash;1982 (2010).\u003c/li\u003e\n\u003cli\u003eShrout, P. E. \u0026amp; Fleiss, J. L. Intraclass Correlations: Uses in Assessing Rater Reliability.\u003cem\u003e \u003c/em\u003e\u003cem\u003ePsychological Bulletin, 86\u003c/em\u003e(2), 420\u0026ndash;428 (1979).\u003c/li\u003e\n\u003cli\u003eAl-Amri, M. \u003cem\u003eet al.\u003c/em\u003e Inertial Measurement Units for Clinical Movement Analysis: Reliability and Concurrent Validity. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 719 (2018).\u003c/li\u003e\n\u003cli\u003eBauer, L., Hamberger, M. A., B\u0026ouml;cker, W., Polzer, H. \u0026amp; Baumbach, S. F. Reliability testing of an IMU-based 2-segment foot model for clinical gait analysis. \u003cem\u003eGait Posture\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 112\u0026ndash;118 (2024).\u003c/li\u003e\n\u003cli\u003eNordin, M. \u0026amp; Frankel, H. \u003cem\u003eBasic Biomechanics of the Musculoskeletal System\u003c/em\u003e (Lippincott Williams \u0026amp; Wilkins, 2012).\u003c/li\u003e\n\u003cli\u003eBauer, L., Hamberger, M. A., B\u0026ouml;cker, W., Polzer, H. \u0026amp; Baumbach, S. F. Development of an IMU based 2-segment foot model for an applicable medical gait analysis. \u003cem\u003eBMC Musculoskelet. Disord.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 606 (2024).\u003c/li\u003e\n\u003cli\u003eSekiguchi, Y. \u003cem\u003eet al.\u003c/em\u003e Evaluation of the Validity, Reliability, and Kinematic Characteristics of Multi-Segment Foot Models in Motion Capture. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 4415 (2020).\u003c/li\u003e\n\u003cli\u003eHu, Q., Liu, L., Mei, F. \u0026amp; Yang, C. Joint Constraints Based Dynamic Calibration of IMU Position on Lower Limbs in IMU-MoCap. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 7161 (2021).\u003c/li\u003e\n\u003cli\u003eEkdahl, M., Loewen, A., Erdman, A., Sahin, S. \u0026amp; Ulman, S. Inertial Measurement Unit Sensor-to-Segment Calibration Comparison for Sport-Specific Motion Analysis. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 7987 (2023).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Optical motion-capture system, inertial measurement unit, ankle kinematics, gait analysis, diverse walking conditions, validity and repeatability","lastPublishedDoi":"10.21203/rs.3.rs-7043663/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7043663/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccurate assessment of three-dimensional ankle kinematics is essential for advancing biomechanical analyses and supporting evidence-based clinical interventions. While optical motion-capture systems (OMCS) are considered the gold standard, their high cost and limited portability restrict their use outside the laboratory. Inertial measurement units (IMUs) offer more practical alternative; however, concerns remain regarding their validity and reliability in diverse walking conditions. This study evaluated the validity and repeatability of IMU-derived ankle kinematics relative to OMCS across three walking surfaces: level ground, inward wedge, and outward wedge. Ankle movements in the sagittal, frontal, and transverse planes were analyzed using a minimal sensor configuration. High agreement and repeatability were observed in the sagittal and transverse planes, particularly during level- and inward-inclined walking. Conversely, the frontal plane demonstrated limited agreement under wedged conditions, likely because of complex multisegmental foot dynamics and reference frame misalignment from static calibration. Despite these limitations, consistent repeatability across conditions supports the use of IMUs for tracking ankle kinematics outside the laboratory. These findings suggest that IMUs are applicable as practical tools for clinical and biomechanical studies, particularly where streamlined setups are warranted.\u003c/p\u003e","manuscriptTitle":"Comparison of 3D ankle kinematics between minimal inertial measurement units configuration and optical motion-capture system under diverse walking conditions","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-15 17:04:44","doi":"10.21203/rs.3.rs-7043663/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"180723536167566087810830259965566262137","date":"2025-07-12T09:08:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"86138878422666705035046262101720021488","date":"2025-07-10T13:29:47+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-10T05:08:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-10T05:06:26+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-04T06:44:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-04T06:43:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-04T06:26:33+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b6636eca-f954-44e3-8777-f695e62a8650","owner":[],"postedDate":"July 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":51433586,"name":"Health sciences/Anatomy"},{"id":51433587,"name":"Physical sciences/Engineering"},{"id":51433588,"name":"Health sciences/Health care"},{"id":51433589,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-02-16T16:07:18+00:00","versionOfRecord":{"articleIdentity":"rs-7043663","link":"https://doi.org/10.1038/s41598-026-39161-8","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-11 15:58:32","publishedOnDateReadable":"February 11th, 2026"},"versionCreatedAt":"2025-07-15 17:04:44","video":"","vorDoi":"10.1038/s41598-026-39161-8","vorDoiUrl":"https://doi.org/10.1038/s41598-026-39161-8","workflowStages":[]},"version":"v1","identity":"rs-7043663","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7043663","identity":"rs-7043663","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}