OpenCap: Human movement dynamics from smartphone videos

Research status and application scenarios of 3D human body modelling methods in the garment ergonomics: a systematic review

This research aims to enhance the comprehension of 3D human modelling methodologies pertinent to the garment ergonomics field. Through a search and analysis of 442 literatures, this study found that, despite the utilisation of high-resolution scanning and sophisticated 3D software, generating the vast diversity of human physiques in models remains a formidable challenge, attributed to issues such as data discrepancies and loss of detail from self-occlusion. Furthermore, through an exhaustive literature survey, this research formulates a framework for juxtaposing various modelling methodologies, analysing their technical tenets, benefits, and limitations from a synergetic and iterative standpoint. Finally, the article underscores future research trajectories, emphasising the critical need to ameliorate model precision and operational efficiency, alongside the integration of garment ergonomics knowledge into 3D human modelling. This research furnishes valuable insights and directions for forthcoming studies, aiming to drive the progression of garment ergonomics towards a more genuine and efficient.

A systematic literature review of computer vision-based biomechanical models for physical workload estimation

Ergonomic risks, driven by strenuous physical demands in complex work settings, are prevalent across industries. Addressing these challenges through detailed assessment and effective interventions enhances safety and employee well-being. Proper and timely measurement of physical workloads is the initial step towards holistic ergonomic control. This study comprehensively explores existing computer vision-based biomechanical analysis methods for workload assessment, assessing their performance against traditional techniques, and categorising them for easier use. Recent strides in artificial intelligence have revolutionised workload assessment, especially in realistic work settings where conventional methods fall short. However, understanding the accuracy, characteristics, and practicality of computer vision-based methods versus traditional approaches remains limited. To bridge this knowledge gap, a literature review along with a meta-analysis was completed in this study to illuminate model accuracy, advantages, and challenges, offering valuable insights for refined technology implementation in diverse work environments.

Markerless Kinematic Data in the Frontal Plane Contributions to Movement Quality in the Single-Leg Squat Test: A Comparison and Decision Tree Approach

OBJECTIVE

The aim of this study is to compare kinematic data of the frontal trunk, pelvis, knee, and summated angles (trunk plus knee) among categorized grades using the single-leg squat (SLS) test, to classify the SLS grade, and to investigate the association between the SLS grade and the frontal angles using smartphone-based markerless motion capture.

METHODS

Ninety-one participants were categorized into 3 grades (good, reduced, and poor) based on the quality of the SLS test. An automated pose estimation algorithm was employed to assess the frontal joint angles during SLS, which were captured by a single smartphone camera. Analysis of variance and a decision tree model using classification and regression tree analysis were utilized to investigate intergroup differences, classify the SLS grades, and identify associations between the SLS grade and frontal angles, respectively.

RESULTS

In the poor group, each frontal trunk, knee, and summated angle was significantly larger than in the good group. Classification and regression tree analysis showed that frontal knee and summated angles could classify the SLS grades with a 76.9% accuracy. Additionally, the classification and regression tree analysis established cutoff points for each frontal knee (11.34°) and summated angles (28.4°), which could be used in clinical practice to identify individuals who have a reduced or poor grade in the SLS test.

CONCLUSIONS

The quality of SLS was found to be associated with interactions among frontal knee and summated angles. With an automated pose estimation algorithm, a single smartphone computer vision method can be utilized to compare and distinguish the quality of SLS movement for remote clinical and sports assessments.

Two cameras can be as good as four for markerless hand tracking during simple finger movements

Recording and quantifying hand and finger movement is essential for understanding the neuromechanical control of the hand. Typically, kinematics are collected through marker-based optoelectronic motion capture systems. However, marker-based systems are time-consuming to setup, expensive, and cumbersome, especially for finger tracking. Advances in markerless systems have potential to overcome these limitations, as demonstrated by recent applications in lower extremity biomechanics research. In this work, we aimed to integrate markerless systems for hand biomechanics research by combining open source markerless motion capture pipelines (MediaPipe and Anipose) and investigating the number of cameras required for tracking single finger flexion-extension movements. Finger movements were recorded at three different speeds (0.50, 0.75, 1 Hz) for each of the instructed fingers (index, middle, ring, little) using 4 webcams. Finger joint angles were compared when using all 4 webcams for triangulating 3D hand key points versus all 2- and 3-camera subset combinations. The number of cameras was found to affect joint angles, with differences up to 20° when using 2 or 3 cameras compared to using all 4 cameras. However, we found some 2-camera orientations had minimal differences compared to using all 4 cameras (< 4° difference for the sum of finger [metacarpal, proximal interphalangeal, and distal phalangeal] joint angles). Thus, there can be little to no benefit of adding more than 2 cameras for 3D markerless tracking of the hand during single finger flexion-extension with optimal camera placement.

Reliability of artificial intelligence-driven markerless motion capture in gait analyses of healthy adults

The KinaTrax markerless motion capture system, used extensively in the analysis of baseball pitching and hitting, is currently being adapted for use in clinical biomechanics. In clinical and laboratory environments, repeatability is inherent to the quality of any diagnostic tool. The KinaTrax system was assessed on within- and between-session reliability for gait kinematic and spatiotemporal parameters in healthy adults. Nine subjects contributed five trials per session over three sessions to yield 135 unique trials. Each trial was comprised of a single bilateral gait cycle. Ten spatiotemporal parameters for each session were calculated and compared using the intraclass correlation coefficient (ICC), Standard Error of the Measurement (SEM), and minimal detectable change (MDC). In addition, seven kinematic waveforms were assessed from each session and compared using the coefficient of multiple determination (CMD). ICCs for between-session spatiotemporal parameters were lowest for left step time (0.896) and left cadence (0.894). SEMs were 0.018 (s) and 3.593 (steps/min) while MDCs were 0.050 (s) and 9.958 (steps/min). Between-session average CMDs for joint angles were large (0.969) in the sagittal plane, medium (0.554) in the frontal plane, and medium (0.327) in the transverse plane while average CMDs for segment angles were large (0.860), large (0.651), and medium (0.561), respectively. KinaTrax markerless motion capture system provides reliable spatiotemporal measures within and between sessions accompanied by reliable kinematic measures in the sagittal and frontal plane. Considerable strides are necessary to improve methodological comparisons, however, markerless motion capture poses a reliable application for gait analysis within healthy individuals.

Difference in movement coordination and variability during Five-Repetition Sit-to-Stand between people with and without Chronic Low back pain

Chronic low back pain (CLBP) affects people's activities of daily living, including sitting down and standing up. Movement pattern analyses during five-repetition sit-to-stand (5RSTS) may allow CLBP status differentiation. 44 CLBP and 22 asymptomatic participants performed 5RSTS in this study, with their trunk and lower limb movements recorded using 3-dimensional motion capture system. Joint active range of motion, joint maximal velocity, joint and segment continuous relative phase (CRP) were analyzed. Mean absolute relative phase (MARP) and deviation phase (DP) variables were calculated in CRP analysis. Between-group kinematic variables were compared using One-way Multivariate Analysis of Covariance (MANCOVA). Significant variables from different methods were compared using binomial logistic regression to assess accuracy for CLBP status. Results showed that segmental CRP is the most sensitive method for CLBP assessment, with the CLBP group femur-to-pelvis and lumbar-to-pelvis movement coordination was more in-phase MARP (F(8,56) = 7.127, p < 0.001, Wilks'Λ = 0.441, ηp2 = 0.559) and stable DP (F(8,56) = 4.585, p < 0.001, Wilks'Λ = 0.551, ηp2 = 0.449) during both standing up and sitting down. Utilizing CRP variables yielded Nagelkerke R2 = 0.708 and overall correct classification of 93 % for CLBP status. Individuals with CLBP exhibited distinct movement coordination and stability, which should be considered in CLBP assessments and intervention. Variable combination from the segment analysis was found to be the most predictive to CLBP status, and significantly different to the results obtained from joint analysis, highlighting the necessity for CRP method standardization in future studies.

Evaluation of a smartphone-based markerless system to measure lower-limb kinematics in patients with knee osteoarthritis

OpenCap, a smartphone-based markerless system, offers a cost-effective alternative to traditional marker-based systems for gait analysis. However, its kinematic measurement accuracy must be evaluated before widespread use in clinical practice. This study aimed to evaluate OpenCap for lower-limb joint angle measurements during walking in patients with knee osteoarthritis (OA) and to compare error metrics between patients and healthy controls. Lower-limb kinematic data were simultaneously collected from 53 patients with knee OA and 30 healthy individuals using OpenCap and a marker-based motion capture system while walking at a self-selected speed. Evaluation was assessed through root mean square error (RMSE) and intraclass correlation coefficient (ICC). Two-way repeated measures analyses of variance were employed to evaluate the main effects of and interactions between group (knee OA patients vs. healthy controls) and walking direction (toward vs. away from the camera). The results demonstrated a grand mean RMSE of 6.1° and an ICC of 0.67 for knee OA patients when walking toward the camera. Patients with knee OA exhibited significantly higher RMSE and lower ICC values compared to healthy controls. Additionally, walking toward the camera was associated with significantly lower RMSE and higher ICC values than walking away from the camera. OpenCap's minimal hardware costs, free software, and user-friendly interface suggest its potential for widespread clinical implementation. The sagittal hip and knee angles demonstrate strong agreement with the marker-based system; however, caution is warranted in clinical decision-making for this population, as errors in most joint angles slightly surpass acceptable thresholds.

Clinical Whole-Body Gait Characterization Using a Single RGB-D Sensor

Instrumented gait analysis is widely used in clinical settings for the early detection of neurological disorders, monitoring disease progression, and evaluating fall risk. However, the gold-standard marker-based 3D motion analysis is limited by high time and personnel demands. Advances in computer vision now enable markerless whole-body tracking with high accuracy. Here, we present vGait, a comprehensive 3D gait assessment method using a single RGB-D sensor and state-of-the-art pose-tracking algorithms. vGait was validated in healthy participants during frontal- and sagittal-perspective walking. Performance was comparable across perspectives, with vGait achieving high accuracy in detecting initial and final foot contacts (F1 scores > 95%) and reliably quantifying spatiotemporal gait parameters (e.g., stride time, stride length) and whole-body coordination metrics (e.g., arm swing and knee angle ROM) at different levels of granularity (mean, step-to-step variability, side asymmetry). The flexibility, accuracy, and minimal resource requirements of vGait make it a valuable tool for clinical and non-clinical applications, including outpatient clinics, medical practices, nursing homes, and community settings. By enabling efficient and scalable gait assessment, vGait has the potential to enhance diagnostic and therapeutic workflows and improve access to clinical mobility monitoring.

Quantifying performance and joint kinematics in functional tasks crucial for anterior cruciate ligament rehabilitation using smartphone video and pose detection

BACKGROUND

The assessment of performance during functional tasks and the quality of movement execution are crucial metrics in the rehabilitation of patients with anterior cruciate ligament (ACL) injuries. While measuring performance is feasible in clinical practice, quantifying joint kinematics poses greater challenges. The aim of this study was to investigate whether smartphone video, using deep neural networks for human pose detection, can enable the clinicians not only to measure performance in functional tasks but also to assess joint kinematics.

METHODS

Twelve healthy participants performed the forward reach of the Star Excursion Balance Test 10 times, along with 10 repetitions of forward jumps and vertical jumps, with simultaneous motion capture via a marker-based reference system and a smartphone. OpenPifPaf was utilized for markerless detection of anatomical landmarks in video recordings. The OpenPifPaf coordinates were scaled using anthropometric data of the thigh, and task performance and joint kinematics were computed for both the marker-based and markerless systems.

RESULTS

Comparing results for marker-based and markerless systems revealed similar joint angles, with mean root mean square errors of 2.8° for the knee, 3.1° for the hip, and 3.9° for the ankle. Excellent agreement was observed for clinically pertinent parameters, i.e., the performance, the peak knee flexion, and the knee range of motion (intraclass correlation coefficient > 0.97).

CONCLUSION

The results underscore the feasibility of using markerless methods based on OpenPifPaf for assessing performance and joint kinematics in functional tasks crucial for ACL patients' rehabilitation. The simplicity of this approach makes it suitable for integration into clinical practice.

Advancing Field-Based Vertical Jump Analysis: Markerless Pose Estimation vs. Force Plates

The countermovement vertical jump (CMJ) is widely used in sports science and rehabilitation to assess lower body power. In controlled laboratory environments, a complex analysis of CMJ performance is usually carried out using motion capture or force plate systems, providing detailed insights into athlete's movement mechanics. While these systems are highly accurate, they are often costly or limited to laboratory settings, making them impractical for widespread or field use. This study aimed to evaluate the accuracy of MMPose, a markerless 2D pose estimation framework, for CMJ analysis by comparing it with force plates. Twelve healthy participants performed five CMJs, with each jump trial simultaneously recorded using force plates and a smartphone camera. Vertical velocity profiles and key temporal variables, including jump phase durations, maximum jump height, vertical velocity, and take-off velocity, were analyzed and compared between the two systems. The statistical methods included a Bland-Altman analysis, correlation coefficients (r), and effect sizes, with consistency and systematic differences assessed using intraclass correlation coefficients (ICC) and paired samples t-tests. The results showed strong agreement (r = 0.992) between the markerless system and force plates, validating MMPose for CMJ analysis. The temporal variables also demonstrated high reliability (ICC > 0.9), with minimal systematic differences and negligible effect sizes for most variables. These findings suggest that the MMPose-based markerless system is a cost-effective and practical alternative for analyzing CMJ performance, particularly in field settings where force plates may be less accessible. This system holds potential for broader applications in sports performance and rehabilitation, enabling more scalable, data-driven movement assessments.

Smartphone videos-driven musculoskeletal multibody dynamics modelling workflow to estimate the lower limb joint contact forces and ground reaction forces

The estimation of joint contact forces in musculoskeletal multibody dynamics models typically requires the use of expensive and time-consuming technologies, such as reflective marker-based motion capture (Mocap) system. In this study, we aim to propose a more accessible and cost-effective solution that utilizes the dual smartphone videos (SPV)-driven musculoskeletal multibody dynamics modeling workflow to estimate the lower limb mechanics. Twelve participants were recruited to collect marker trajectory data, force plate data, and motion videos during walking and running. The smartphone videos were initially analyzed using the OpenCap platform to identify key joint points and anatomical markers. The markers were used as inputs for the musculoskeletal multibody dynamics model to calculate the lower limb joint kinematics, joint contact forces, and ground reaction forces, which were then evaluated by the Mocap-based workflow. The root mean square error (RMSE), mean absolute deviation (MAD), and Pearson correlation coefficient (ρ) were adopted to evaluate the results. Excellent or strong Pearson correlations were observed in most lower limb joint angles (ρ = 0.74 ~ 0.94). The averaged MADs and RMSEs for the joint angles were 1.93 ~ 6.56° and 2.14 ~ 7.08°, respectively. Excellent or strong Pearson correlations were observed in most lower limb joint contact forces and ground reaction forces (ρ = 0.78 ~ 0.92). The averaged MADs and RMSEs for the joint lower limb joint contact forces were 0.18 ~ 1.07 bodyweight (BW) and 0.28 ~ 1.32 BW, respectively. Overall, the proposed smartphone video-driven musculoskeletal multibody dynamics simulation workflow demonstrated reliable accuracy in predicting lower limb mechanics and ground reaction forces, which has the potential to expedite gait dynamics analysis in a clinical setting.

Osteoarthritis year in review 2024: Biomechanics

OBJECTIVE

We aimed to systematically review and summarize the literature of the past year on osteoarthritis (OA) and biomechanics, to highlight gaps and challenges, and to present some promising approaches and developments.

METHODS

A systematic literature search was conducted using Pubmed and the Web of Science Core Collection. We included original articles and systematic reviews on OA and biomechanics in human subjects published between April 2023 and April 2024.

RESULTS

Of the 155 studies that met the inclusion criteria, 9 were systematic reviews and 146 were original (mostly cross-sectional) studies that included a total of 6488 patients and 1921 controls with a mean age of 57.5 and 44.7 years, respectively. Promising advances have been made in medical imaging of affected soft tissue structures, the relationship between soft tissue properties and biomechanical changes in OA, new technologies to facilitate easier assessment of ambulatory biomechanics, and personalized physics-based models that also include complex chemical and mechanobiological mechanisms, all of which are relevant to gaining mechanistic insights into the pathophysiology of OA.

CONCLUSIONS

There is still an unmet need for larger longitudinal data sets that combine clinical, radiological, and biomechanical outcomes to characterize the biomechanical fingerprint that underlies the trajectory of functional decline and biomechanical phenotypes of OA. In addition, criteria and guidelines for control groups, as well as methods and standards for model verification allowing for comparisons between studies are needed.

Open Source, Open Science: Development of OpenLESS as the Automated Landing Error Scoring System

Context

The Open Landing Error Scoring System (OpenLESS) is a novel development aimed at automating the LESS for assessment of lower extremity movement quality during a jump-landing task. With increasing utilization of clinical measures to monitor outcomes and limited time during clinical visits for a lengthy analysis of functional movement, there is a pressing need to extend automation efforts. Addressing these issues, OpenLESS is an open-source tool that utilizes a freely available markerless motion capture system to automate the LESS using three-dimensional kinematics.

Objective

To describe the development of OpenLESS, examine the validity against expert rater LESS scores in healthy and clinically relevant cohorts, and assess the intersession reliability collected across four time points in an athlete cohort.

Design

Observational.

Participants

92 participants (72 females and 20 males, mean age 23.3 years) from healthy, post-anterior cruciate ligament reconstruction (ACLR; median 33 months since surgery), and amateur athlete cohorts.

Main Outcome Measures

A software package, "OpenLESS," was developed to interpret movement quality (LESS score) from kinematics captured from markerless motion capture. Validity and reliability were assessed with intraclass correlation coefficients (ICC), standard error of measure (SEM), and minimal detectable change (MDC).

Results

OpenLESS agreed well with expert rater LESS scores for healthy (ICC 2, k =0.79) and clinically relevant, post-ACLR cohorts (ICC 2, k =0.88). The automated OpenLESS system reduced scoring time, processing all 159 trials in under 15 minutes compared to the 18.5 hours (7 minutes per trial) required for manual expert rater scoring. When tested outside laboratory conditions, OpenLESS showed excellent reliability across repeated sessions (ICC 2, k >0.89), with a SEM of 0.98 errors and MDC of 2.72 errors.

Conclusion

OpenLESS shows promise as an efficient, automated tool for clinically assessing jump-landing quality, with good validity versus experts in healthy and post-ACLR populations, and excellent field reliability, addressing the need for objective movement analysis.

KEY POINTS

OpenLESS accurately detected jump-landing events (ICC>0.99) using markerless motion capture, validating its use as an alternative to laboratory-based force plate measurements.The automated scoring system showed good agreement with expert raters in healthy (ICC=0.79) and post-ACLR (ICC=0.88) populations.OpenLESS demonstrated good to excellent test-retest reliability (ICC=0.89) across multiple testing sessions, with minimal score variation, supporting its utility for longitudinal movement assessment.

Synchronised Video, Motion Capture and Force Plate Dataset for Validating Markerless Human Movement Analysis

The BioCV dataset is a unique combination of synchronised multi-camera video, marker based optical motion capture, and force plate data, observing 15 healthy participants (7 males, 8 females) performing controlled and repeated motions (walking, running, jumping and hopping), as well as photogrammetry scan data for each participant. The dataset was created for the purposes of developing and validating the performance of computer vision based markerless motion capture systems with respect to marker based systems.

Using a Webcam to Assess Upper Extremity Proprioception: Experimental Validation and Application to Persons Post Stroke

Many medical conditions impair proprioception but there are few easy-to-deploy technologies for assessing proprioceptive deficits. Here, we developed a method-called "OpenPoint"-to quantify upper extremity (UE) proprioception using only a webcam as the sensor. OpenPoint automates a classic neurological test: the ability of a person to use one hand to point to a finger on their other hand with vision obscured. Proprioception ability is quantified with pointing error in the frontal plane measured by a deep-learning-based, computer vision library (MediaPipe). In a first experiment with 40 unimpaired adults, pointing error significantly increased when we replaced the target hand with a fake hand, verifying that this task depends on the availability of proprioceptive information from the target hand, and that we can reliably detect this dependence with computer vision. In a second experiment, we quantified UE proprioceptive ability in 16 post-stroke participants. Individuals post stroke exhibited increased pointing error (p < 0.001) that was correlated with finger proprioceptive error measured with an independent, robotic assessment (r = 0.62, p = 0.02). These results validate a novel method to assess UE proprioception ability using affordable computer technology, which provides a potential means to democratize quantitative proprioception testing in clinical and telemedicine environments.

Role of Thigh Muscle Strength and Joint Kinematics in Dynamic Stability: Implications for Y-Balance Test Performance

CONTEXT

The Y-Balance Test Lower Quarter (YBT-LQ) is a widely utilized tool for evaluating dynamic postural control, requiring a combination of mobility and strength. This study aimed to investigate the combined relationship between isometric thigh muscle strength and joint kinematics on YBT-LQ performance.

DESIGN

Cross-sectional laboratory study.

METHODS

Isometric quadriceps and hamstrings strength were measured before the YBT-LQ in 39 healthy participants (27 females and 12 males). The test was performed under 3-dimensional markerless motion capture, where joint kinematics were extracted from the maximum reach position from each direction. Three multivariable linear regression models were then used to determine the strongest combination of predictors for YBT-LQ performance.

RESULTS

Greater hamstrings strength and increased knee flexion, ankle dorsiflexion, and trunk ipsilateral-flexion joint angles explained 56.8% (P < .001) of the variance in anterior reach. Hip flexion, knee flexion, and ankle dorsiflexion angles were the strongest predictors for posteromedial reach distance, explaining 73.0% of the variance (P < .001). Last, 43.3% (P < .001) of the variance in posterolateral reach distance was predicted by hamstring strength and knee-flexion angle.

CONCLUSIONS

These results emphasize the importance of hamstring strength in YBT-LQ performance across different reach directions. Additionally, the kinematics illustrate a potential movement strategy for maximizing reach distance on the YBT-LQ in healthy individuals. Clinicians can utilize this information to guide interventions aimed at improving dynamic postural control, particularly by focusing on increasing hamstring strength and testing for impairments in specific movement patterns.

Identification of distinct subtypes of post-stroke and neurotypical gait behaviors using neural network analysis of kinematic time series data

Background

Heterogeneous types of gait impairment are common post-stroke. Studies used supervised and unsupervised machine learning on discrete biomechanical features to summarize the gait cycle and identify common patterns of gait behaviors. However, discrete features cannot account for temporal variations that occur during gait. Here, we propose a novel machine-learning pipeline to identify subgroups of gait behaviors post-stroke using kinematic time series data.

Methods

We analyzed ankle and knee kinematic data during treadmill walking data in 39 individuals post-stroke and 28 neurotypical controls. The data were first input into a supervised dual-stage Convolutional Neural Network-Temporal Convolutional Network, trained to extract temporal and spatial gait features. Then, we used these features to find clusters of different gait behaviors using unsupervised time series k-means. We repeated the clustering process using 10,000 bootstrap training data samples and a Gaussian Mixture Model to identify stable clusters representative of our dataset. Finally, we assessed the kinematic differences between the identified clusters using 1D statistical parametric mapping ANOVA. We then compared gait spatiotemporal and clinical characteristics between clusters using one-way ANOVA.

Results

We obtained five clusters: two clusters of neurotypical individuals (C1 and C2) and three clusters of individuals post-stroke (S1, S2, S3). C1 had kinematics that resembled the normative gait pattern. Individuals in C2 had a shorter stride time than C1. Individuals in S1 had mild impairment and walked with increased bilateral knee flexion during the loading response. Individuals in S2 had moderate impairment, were the slowest among the clusters, took shorter steps, had increased knee flexion during stance bilaterally and reduced paretic knee flexion during swing. Individuals in S3 had mild impairment, asymmetric swing time, had increased ankle abduction during the gait cycle and reduced dorsiflexion bilaterally during loading response and stance. Every individual was assigned to a cluster with a cluster membership likelihood above 93%.

Conclusions

Our results indicate that joint kinematics in individuals post-stroke are distinct from controls, even in those individuals with mild impairment. The three subgroups post-stroke showed distinct kinematic impairments during specific phases in the gait cycle, providing additional information to clinicians for gait retraining interventions.

Evaluation of drop vertical jump kinematics and kinetics using 3D markerless motion capture in a large cohort

Introduction

3D Markerless motion capture technologies have advanced significantly over the last few decades to overcome limitations of marker-based systems, which require significant cost, time, and specialization. As markerless motion capture technologies develop and mature, there is increasing demand from the biomechanics community to provide kinematic and kinetic data with similar levels of reliability and accuracy as current reference standard marker-based 3D motion capture methods. The purpose of this study was to evaluate how a novel markerless system trained with both hand-labeled and synthetic data compares to lower extremity kinematic and kinetic measurements from a reference marker-based system during the drop vertical jump (DVJ) task.

Methods

Synchronized video data from multiple camera views and marker-based data were simultaneously collected from 127 participants performing three repetitions of the DVJ. Lower limb joint angles and joint moments were calculated and compared between the markerless and marker-based systems. Root mean squared error values and Pearson correlation coefficients were used to quantify agreement between the systems.

Results

Root mean squared error values of lower limb joint angles and joint moments were ≤ 9.61 degrees and ≤ 0.23 N×m/kg, respectively. Pearson correlation values between markered and markerless systems were 0.67-0.98 hip, 0.45-0.99 knee and 0.06-0.99 ankle for joint kinematics. Likewise, Pearson correlation values were 0.73-0.90 hip, 0.61-0.95 knee and 0.74-0.95 ankle for joint kinetics.

Discussion

These results highlight the promising potential of markerless motion capture, particularly for measures of hip, knee and ankle rotations. Further research is needed to evaluate the viability of markerless ankle measures in the frontal plane to determine if differences in joint solvers are inducing unanticipated error.

Validation of Automated Countermovement Vertical Jump Analysis: Markerless Pose Estimation vs. 3D Marker-Based Motion Capture System

This study aimed to validate the automated temporal analysis of countermovement vertical jump (CMJ) using MMPose, a markerless pose estimation framework, by comparing it with the gold-standard 3D marker-based motion capture system. Twelve participants performed five CMJ trials, which were simultaneously recorded using the marker-based system and two smartphone cameras capturing both sides of the body. Key kinematic points, including center of mass (CoM) and toe trajectories, were analyzed to determine jump phases and temporal variables. The agreement between methods was assessed using Bland-Altman analysis, root mean square error (RMSE), and Pearson's correlation coefficient (r), while consistency was evaluated via intraclass correlation coefficient (ICC 3,1) and two-way repeated-measures ANOVA. Cohen's effect size (d) quantified the practical significance of differences. Results showed strong agreement (r > 0.98) with minimal bias and narrow limits of agreement for most variables. The markerless system slightly overestimated jump height and CoM vertical velocity, but ICC values (ICC > 0.91) confirmed strong reliability. Cohen's d values were near zero, indicating trivial differences, and no variability due to recording side was observed. Overall, MMPose proved to be a reliable alternative for in-field CMJ analysis, supporting its broader application in sports and rehabilitation settings.

Ethnokinesiology: towards a neuromechanical understanding of cultural differences in movement

Each individual's movements are sculpted by constant interactions between sensorimotor and sociocultural factors. A theoretical framework grounded in motor control mechanisms articulating how sociocultural and biological signals converge to shape movement is currently missing. Here, we propose a framework for the emerging field of ethnokinesiology aiming to provide a conceptual space and vocabulary to help bring together researchers at this intersection. We offer a first-level schema for generating and testing hypotheses about cultural differences in movement to bridge gaps between the rich observations of cross-cultural movement variations and neurophysiological and biomechanical accounts of movement. We explicitly dissociate two interacting feedback loops that determine culturally relevant movement: one governing sensorimotor tasks regulated by neural signals internal to the body, the other governing ecological tasks generated through actions in the environment producing ecological consequences. A key idea is the emergence of individual-specific and culturally influenced motor concepts in the nervous system, low-dimensional functional mappings between sensorimotor and ecological task spaces. Motor accents arise from perceived differences in motor concept topologies across cultural contexts. We apply the framework to three examples: speech, gait and grasp. Finally, we discuss how ethnokinesiological studies may inform personalized motor skill training and rehabilitation, and challenges moving forward.This article is part of the theme issue 'Minds in movement: embodied cognition in the age of artificial intelligence'.

Repeatability and minimal detectable change including clothing effects for smartphone-based 3D markerless motion capture

OpenCap, a smartphone- and web-based markerless system, has shown acceptable accuracy compared to marker-based systems, but lacks information on repeatability. This study fills this gap by evaluating the intersession repeatability of OpenCap and investigating the effects of clothing on gait kinematics. Twenty healthy volunteers participated in a test-retest study, performing walking and sit-to-stand tasks with minimal clothing and regular street wear. Segment lengths and lower-limb kinematics were compared between both sessions and for both clothing conditions using the root-mean-square-deviation (RMSD) for entire waveforms and the standard error of measurement (SEM) and minimal detectable change (MDC) for discrete kinematic parameters. In general, the RMSD test-retest values were 2.8 degrees (SD: 1.0) for walking and 3.3 degrees (SD: 1.2) for sit-to-stand. The highest intersession variability was observed in the trunk, pelvis, and hip kinematics of the sagittal plane. SEM and MDC values were on average 2.2 and 6.0 degrees, respectively, for walking, and 2.4 and 6.5 degrees for sit-to-stand. Clothing had minimal effects on kinematics by adding on average less than one degree to the RMSD values for most variables. The segment lengths showed moderate to excellent agreement between both sessions and poor to moderate agreement between clothing conditions. The study highlights the reliability of OpenCap for markerless motion capture, emphasizing its potential for large-scale field studies. However, some variables showed high MDC values above 5 degrees and thus warrant further enhancement of the technology. Although clothing had minimal effects, it is still recommended to maintain consistent clothing to minimize overall variability.

Fatigue-Related Changes in Running Technique and Mechanical Variables After a Maximal Incremental Test in Recreational Runners

Understanding the changes in running mechanics caused by fatigue is essential to assess its impact on athletic performance. Changes in running biomechanics after constant speed conditions are well documented, but the adaptive responses after a maximal incremental test are unknown. We compared the spatiotemporal, joint kinematics, elastic mechanism, and external work parameters before and after a maximal incremental treadmill test. Eighteen recreational runners performed 2-minute runs at 8 km·h-1 before and after a maximal incremental test on a treadmill. Kinematics, elastic parameters, and external work were determined using the OpenCap and OpenSim software. We did not find differences in spatiotemporal parameters and elastic parameters (mechanical work, ankle, and knee motion range) between premaximal and postmaximal test conditions. After the maximal test, the runners flexed their hips more at contact time (19.4°-20.6°, P = .013) and presented a larger range of pelvis rotation at the frontal plane (10.3°-11.4°, P = .002). The fatigue applied in the test directly affects pelvic movements; however, it does not change the lower limb motion or the spatiotemporal and mechanical work parameters in recreational runners. A larger frontal plane motion of the pelvis deserves attention due to biomechanical risk factors associated with injuries.

Towards Wearable and Portable Spine Motion Analysis Through Dynamic Optimization of Smartphone Videos and IMU Data

BACKGROUND

Monitoring spine kinematics is crucial for applications like disease evaluation and ergonomics analysis. However, the small scale of vertebrae and the number of degrees of freedom present significant challenges for noninvasive and convenient spine kinematics estimation.

METHODS

This study developed a dynamic optimization framework for wearable spine motion tracking at the intervertebral joint level by integrating smartphone videos and Inertia Measurement Units (IMUs) with dynamic constraints from a thoracolumbar spine model. Validation involved motion data from 10 healthy males performing static standing, dynamic upright trunk rotations, and gait. This data included rotations of ten IMUs on vertebrae and virtual landmarks from three smartphone videos preprocessed by OpenCap, an application leveraging computer vision for pose estimation. The kinematic measures derived from the optimized solution were compared against simultaneously collected infrared optical marker-based measurements and in vivo literature data. Solutions only based on IMUs or videos were also compared for accuracy evaluation.

RESULTS

The proposed optimization approach closely matched the reference data in the intervertebral or segmental rotation range, demonstrating minimal angular differences across all motions and the highest correlation in 3D rotations (maximal Pearson and intraclass correlation coefficients of 0.92 and 0.94, respectively). Time-series changes of joint angles also aligned well with the optical-marker reference.

CONCLUSION

Dynamic optimization of the spine simulation that integrates IMUs and computer vision outperforms the single-modality method.

SIGNIFICANCE

This markerless 3D spine motion capture method holds potential for spinal health assessment in large cohorts in real-world settings without dedicated laboratories.

Validity and reliability of trunk and lower-limb kinematics during squatting, hopping, jumping and side-stepping using OpenCap markerless motion capture application

OpenCap is a web-based markerless motion capture platform that estimates 3D kinematics from videos recorded from at least two iOS devices. This study aimed to determine the concurrent validity and inter-session reliability of OpenCap for measuring trunk and lower-limb kinematics during squatting, hopping, countermovement jumping, and cutting. Nineteen participants (10 males, 9 females; age 27.7 ± 4.1 years) were included. Countermovement jump, single-leg triple vertical hop, single-leg squat, sidestep cutting and side hop tasks were assessed. For validity, OpenCap was compared to a marker-based motion capture system using root-mean-square error. Test-retest reliability of OpenCap was determined using intraclass correlations and minimum detectable change (MDC) from two testing sessions. The squat had the lowest RMSE across joint angles (mean = 7.0°, range = 2.9° to 13.6°). For peak angles, the countermovement jump (jump phase) (ICC = 0.62-0.93) and the squat (ICC = 0.60-0.92) had the best reliability across all joints. For initial contact, the side hop had the best inter-session reliability (ICC = 0.70-0.94) across all joint angles. As such, OpenCap validity and reliability are joint and task specific.

Real-Time Calibration-Free Musculotendon Kinematics for Neuromusculoskeletal Models

Neuromusculoskeletal (NMS) models enable non-invasive estimation of clinically important internal biomechanics. A critical part of NMS modelling is the estimation of musculotendon kinematics, which comprise musculotendon unit lengths, moment arms, and lines of action. Musculotendon kinematics, which are partially dependent on joint angles, define the non-linear mapping of muscle forces to joint moments and contact forces. Currently, real-time computation of musculotendon kinematics requires creation of a per-individual surrogate model. The computational speed and accuracy of these surrogates degrade with increasing number of coordinates. We developed a feed-forward neural network that completely encodes musculotendon kinematics of a target model across a wide anthropometric range, enabling accurate real-time estimates of musculotendon kinematics without need for a priori creation of a per-individual surrogate model. Compared to reference, the neural network had median normalized errors ~0.1% for musculotendon lengths, <0.4% for moment arms, and <0.10° for line of action orientations. The neural network was employed within an electromyogram-informed NMS model to calculate hip contact forces, demonstrating little difference (normalized root mean square error 1.23±0.15 %) compared to using reference musculotendon kinematics. Finally, execution time was <0.04 ms per frame and constant for increasing number of model coordinates. Our approach to musculoskeletal kinematics may facilitate deployment of complex real-time NMS modelling in computer vision or wearable sensors applications to realize biomechanics monitoring, rehabilitation, and disease management outside the research laboratory.

Biomechanical Gait Analysis Using a Smartphone-Based Motion Capture System (OpenCap) in Patients with Neurological Disorders

This study evaluates the utility of OpenCap (v0.3), a smartphone-based motion capture system, for performing gait analysis in patients with neurological disorders. We compared kinematic and kinetic gait parameters between 10 healthy controls and 10 patients with neurological conditions, including stroke, Parkinson's disease, and cerebral palsy. OpenCap captured 3D movement dynamics using two smartphones, with data processed through musculoskeletal modeling. The key findings indicate that the patient group exhibited significantly slower gait speeds (0.67 m/s vs. 1.10 m/s, p = 0.002), shorter stride lengths (0.81 m vs. 1.29 m, p = 0.001), and greater step length asymmetry (107.43% vs. 91.23%, p = 0.023) compared to the controls. Joint kinematic analysis revealed increased variability in pelvic tilt, hip flexion, knee extension, and ankle dorsiflexion throughout the gait cycle in patients, indicating impaired motor control and compensatory strategies. These results indicate that OpenCap can effectively identify significant gait differences, which may serve as valuable biomarkers for neurological disorders, thereby enhancing its utility in clinical settings where traditional motion capture systems are impractical. OpenCap has the potential to improve access to biomechanical assessments, thereby enabling better monitoring of gait abnormalities and informing therapeutic interventions for individuals with neurological disorders.

Evolving cartilage strain with pain progression and gait: a longitudinal study post-ACL reconstruction at six and twelve months

Background

Anterior cruciate ligament (ACL) injuries are prevalent musculoskeletal conditions often resulting in long-term degenerative outcomes such as osteoarthritis (OA). Despite surgical advances in ACL reconstruction, a significant number of patients develop OA within ten years post-surgery, providing a patient population that may present early markers of cartilage degeneration detectable using noninvasive imaging.

Purpose

This study aims to investigate the temporal evolution of cartilage strain and relaxometry post-ACL reconstruction using displacement under applied loading MRI and quantitative MRI. Specifically, we examined the correlations between MRI metrics and pain, as well as knee loading patterns during gait, to identify early candidate markers of cartilage degeneration.

Materials and Methods

Twenty-five participants (female/male = 15/10; average age = 25.6 yrs) undergoing ACL reconstruction were enrolled in a prospective longitudinal cohort study between 2022 and 2023. MRI scans were conducted at 6- and 12-months post-surgery, assessing T2, T2*, and T1ρ relaxometry values, and intratissue cartilage strain. Changes in pain were evaluated using standard outcome scores, and gait analysis assessed the knee adduction moment (KAM). Regressions were performed to evaluate relationships between MRI metrics in cartilage contact regions, patient-reported pain, and knee loading metrics.

Results

Increases in axial and transverse strains in the tibial cartilage were significantly correlated with increased pain, while decreases in shear strain were associated with increased pain. Changes in strain metrics were also significantly related to KAM at12 months.

Conclusions

Changes in cartilage strain and relaxometry are related to heightened pain and altered knee loading patterns, indicating potential early markers of osteoarthritis progression. These findings underscore the importance of using advanced MRI for early monitoring in ACL-reconstructed patients to optimize treatment outcomes, while also highlighting KAM as a modifiable intervention through gait retraining that may positively impact the evolution of cartilage health and patient pain.

Key Results

Increased axial and transverse strains in the tibial cartilage from 6 to 12 months post-ACL reconstruction were significantly correlated with increased pain, suggesting evolving changes in cartilage biomechanical properties over time.Decreases in shear strain in inner femoral and central tibial cartilage regions were linked to increased pain, indicating alterations in joint loading patterns.Decreases in shear strain in the inner femoral cartilage were significantly associated with decreased 12-month knee adduction moment (KAM), a surrogate for medial cartilage knee loading during walking.

Optimizing Technical Training for Wheelchair-User Billiard Players through Modified Equipment Implementation

This study aims to enhance the effectiveness of the preparation process and the performance of wheelchair users in international billiard competitions through modified equipment. The research methods include analysis and synthesis of the scientific and methodological literature, sociological research methods (questionnaires), expert assessment methods, pedagogical research methods (observation, testing, experimentation), and methods of mathematical statistics. The results of our study are significant: Implementing our developed training program for billiards players with musculoskeletal disorders, utilizing the modified equipment (special mechanical bridge and straps for cue fixation during shots) we created in a pedagogical experiment, demonstrated a probable improvement of 36% in the technical and tactical preparedness of the athletes compared to previous years. This led to a 33% increase in players from the Ukrainian team's competition performance at the national and European Pool Championships (wheelchair division). In conclusion, implementing our developed training program, accompanied by specialized auxiliary equipment, demonstrated promising results in a pedagogical experiment. These findings underscore the potential of the modified equipment and tailored training programs to optimize sports training for individuals with musculoskeletal impairments in adaptive billiards, contributing to the continued humanization of the sport and offering an effective preparation process for the athletes.

Position paper on how technology for human motion analysis and relevant clinical applications have evolved over the past decades: Striking a balance between accuracy and convenience

BACKGROUND

Over the past decades, tremendous technological advances have emerged in human motion analysis (HMA).

RESEARCH QUESTION

How has technology for analysing human motion evolved over the past decades, and what clinical applications has it enabled?

METHODS

The literature on HMA has been extensively reviewed, focusing on three main approaches: Fully-Instrumented Gait Analysis (FGA), Wearable Sensor Analysis (WSA), and Deep-Learning Video Analysis (DVA), considering both technical and clinical aspects.

RESULTS

FGA techniques relying on data collected using stereophotogrammetric systems, force plates, and electromyographic sensors have been dramatically improved providing highly accurate estimates of the biomechanics of motion. WSA techniques have been developed with the advances in data collection at home and in community settings. DVA techniques have emerged through artificial intelligence, which has marked the last decade. Some authors have considered WSA and DVA techniques as alternatives to "traditional" HMA techniques. They have suggested that WSA and DVA techniques are destined to replace FGA.

SIGNIFICANCE

We argue that FGA, WSA, and DVA complement each other and hence should be accounted as "synergistic" in the context of modern HMA and its clinical applications. We point out that DVA techniques are especially attractive as screening techniques, WSA methods enable data collection in the home and community for extensive periods of time, and FGA does maintain superior accuracy and should be the preferred technique when a complete and highly accurate biomechanical data is required. Accordingly, we envision that future clinical applications of HMA would favour screening patients using DVA in the outpatient setting. If deemed clinically appropriate, then WSA would be used to collect data in the home and community to derive relevant information. If accurate kinetic data is needed, then patients should be referred to specialized centres where an FGA system is available, together with medical imaging and thorough clinical assessments.

Validity and Test-Retest Reliability of Spatiotemporal Running Parameter Measurement Using Embedded Inertial Measurement Unit Insoles

Running is the basis of many sports and has highly beneficial effects on health. To increase the understanding of running, DSPro® insoles were developed to collect running parameters during tasks. However, no validation has been carried out for running gait analysis. The aims of this study were to assess the test-retest reliability and criterion validity of running gait parameters from DSPro® insoles compared to a motion-capture system. Equipped with DSPro® insoles, a running gait analysis was performed on 30 healthy participants during overground and treadmill running using a motion-capture system. Using an intraclass correlation coefficient (ICC), the criterion validity and test-retest reliability of spatiotemporal parameters were calculated. The test-retest reliability shows moderate to excellent ICC values (ICC > 0.50) except for propulsion time during overground running at a fast speed with the motion-capture system. The criterion validity highlights a validation of running parameters regardless of speeds (ICC > 0.70). This present study validates the good criterion validity and test-retest reliability of DSPro® insoles for measuring spatiotemporal running gait parameters. Without the constraints of a 3D motion-capture system, such insoles seem to be helpful and relevant for improving the care management of active patients or following running performance in sports contexts.

Markerless motion analysis to assess reaching-sideways in individuals with dyskinetic cerebral palsy: A validity study

This study aimed to evaluate clinical utility of 2D-markerless motion analysis (2DMMA) from a single camera during a reaching-sideways-task in individuals with dyskinetic cerebral palsy (DCP) by determining (1) concurrent validity by correlating 2DMMA against marker-based 3D-motion analysis (3DMA) and (2) construct validity by assessing differences in 2DMMA features between DCP and typically developing (TD) peers. 2DMMA key points were tracked from frontal videos of a single camera by DeepLabCut and accuracy was assessed against human labelling. Shoulder, elbow and wrist angles were calculated from 2DMMA and 3DMA (as gold standard) and correlated to assess concurrent validity. Additionally, execution time and variability features such as mean point-wise standard deviation of the angular trajectories (i.e. shoulder elevation, elbow and wrist flexion/extension) and wrist trajectory deviation by mean overshoot and convex hull were calculated from key points. 2DMMA features were compared between the DCP group and TD peers to assess construct validity. Fifty-one individuals (30 DCP;21 TD; age:5-24 years) participated. An accuracy of approximately 1.5 cm was reached for key point tracking. While significant correlations were found for wrist (ρ = 0.810;p < 0.001) and elbow angles (ρ = 0.483;p < 0.001), 2DMMA shoulder angles were not correlated (ρ = 0.247;p = 0.102) to 3DMA. Wrist and elbow angles, execution time and variability features all differed between groups (Effect sizes 0.35-0.81;p < 0.05). Videos of a reaching-sideways-task processed by 2DMMA to assess upper extremity movements in DCP showed promising validity. The method is especially valuable to assess movement variability.

Marker Data Enhancement For Markerless Motion Capture

Objective

Human pose estimation models can measure movement from videos at a large scale and low cost; however, open-source pose estimation models typically detect only sparse keypoints, which leads to inaccurate joint kinematics. OpenCap, a freely available service for researchers to measure movement from videos, addresses this issue using a deep learning model-the marker enhancer-that transforms sparse keypoints into dense anatomical markers. However, OpenCap performs poorly on movements not included in the training data. Here, we create a much larger and more diverse training dataset and develop a more accurate and generalizable marker enhancer.

Methods

We compiled marker-based motion capture data from 1176 subjects and synthesized 1433 hours of keypoints and anatomical markers to train the marker enhancer. We evaluated its accuracy in computing kinematics using both benchmark movement videos and synthetic data representing unseen, diverse movements.

Results

The marker enhancer improved kinematic accuracy on benchmark movements (mean error: 4.1°, max: 8.7°) compared to using video keypoints (mean: 9.6°, max: 43.1°) and OpenCap's original enhancer (mean: 5.3°, max: 11.5°). It also better generalized to unseen, diverse movements (mean: 4.1°, max: 6.7°) than OpenCap's original enhancer (mean: 40.4°, max: 252.0°).

Conclusion

Our marker enhancer demonstrates both accuracy and generalizability across diverse movements.

Significance

We integrated the marker enhancer into OpenCap, thereby offering its thousands of users more accurate measurements across a broader range of movements.

Calibrationless monocular vision musculoskeletal simulation during gait

With computer vision technology and prediction of ground reaction forces (GRF), a previous study performed markerless motion capture and musculoskeletal simulation with two smartphones (OpenCap). A recent approach can reconstruct 3D human motion from a single video without calibration and it may further simplify the motion capture process. However it has not been combined with musculoskeletal simulation and the validity is unclear. Therefore, the purpose of this study was to determine the validity of the musculoskeletal simulation using a monocular vision approach. An open-source dataset that contains motion capture and video data during gait from 10 healthy participants was used. Human motion reconstruction with the skinned human (SMPL) model was performed on each video. Virtual marker data was generated by extracting the position data from the SMPL skin vertices. Inverse kinematics, GRF prediction (only for monocular vision approach), inverse dynamics and static optimization were performed using a musculoskeletal model for experimental motion capture data and the generated virtual markers from videos. Mean absolute errors (MAE) between motion capture based and monocular vision based simulation outcomes were calculated. The MAE were 8.4° for joint angles, 5.0 % bodyweight for GRF, 1.1 % bodyweight*height for joint moments and 0.11 for estimated muscle activations from 16 muscles. The entire MAE was larger but some were comparable to OpenCap. Using the monocular vision approach, motion capture and musculoskeletal simulation can be done with no preparations and is beneficial for clinicians to quantify the daily gait assessment.

Asymmetry in kinematics of dominant/nondominant lower limbs in central and lateral positioned college and sub-elite soccer players

Change of direction, stops, and pivots are among the most common non-contact movements associated with anterior cruciate ligament (ACL) injuries in soccer. By observing these dynamic movements, clinicians recognize abnormal kinematic patterns that contribute to ACL tears such as increased knee valgus or reduced knee flexion. Different motions and physical demands are observed across playing positions, which may result in varied lower limb kinematic patterns. In the present study, 28 college and sub-elite soccer players performed four dynamic motions (change of direction with and without ball, header, and instep kick) with the goal of examining the effect of on-field positioning, leg dominance, and gender in lower body kinematics. Motion capture software monitored joint angles in the knee, hip, and ankle. A three-way ANOVA showed significant differences in each category. Remarkably, centrally positioned players displayed significantly greater knee adduction (5° difference, p = 0.013), hip flexion (9° difference, p = 0.034), hip adduction (7° difference, p = 0.016), and dorsiflexion (12° difference, p = 0.022) when performing the instep kick in comparison to their laterally positioned counterparts. These findings suggest that central players tend to exhibit a greater range of motion when performing an instep kicking task compared to laterally positioned players. At a competitive level, this discrepancy could potentially lead to differences in lower limb muscle development among on-field positions. Accordingly, it is suggested to implement position-specific prevention programs to address these asymmetries in lower limb kinematics, which can help mitigate dangerous kinematic patterns and consequently reduce the risk of ACL injury in soccer players.

Prediction of Achilles Tendon Force During Common Motor Tasks From Markerless Video

Remodeling of the Achilles tendon (AT) is partly driven by its mechanical environment. AT force can be estimated with neuromusculoskeletal (NMSK) modeling; however, the complex experimental setup required to perform the analyses confines use to the laboratory. We developed task-specific long short-term memory (LSTM) neural networks that employ markerless video data to predict the AT force during walking, running, countermovement jump, single-leg landing, and single-leg heel rise. The task-specific LSTM models were trained on pose estimation keypoints and corresponding AT force data from 16 subjects, calculated via an established NMSK modeling pipeline, and cross-validated using a leave-one-subject-out approach. As proof-of-concept, new motion data of one participant was collected with two smartphones and used to predict AT forces. The task-specific LSTM models predicted the time-series AT force using synthesized pose estimation data with root mean square error (RMSE) ≤ 526 N, normalized RMSE (nRMSE) ≤ 0.21 , R 2 ≥ 0.81 . Walking task resulted the most accurate with RMSE = 189±62 N; nRMSE = 0.11±0.03 , R 2 = 0.92±0.04 . AT force predicted with smartphones video data was physiologically plausible, agreeing in timing and magnitude with established force profiles. This study demonstrated the feasibility of using low-cost solutions to deploy complex biomechanical analyses outside the laboratory.

Inclusion of a skeletal model partly improves the reliability of lower limb joint angles derived from a markerless depth camera

A single depth camera provides a fast and easy approach to performing biomechanical assessments in a clinical setting; however, there are currently no established methods to reliably determine joint angles from these devices. The primary aim of this study was to compare joint angles as well as the between-day reliability of direct kinematics to model-constrained inverse kinematics recorded using a single markerless depth camera during a range of clinical and athletic movement assessments.A secondary aim was to determine the minimum number of trials required to maximize reliability. Eighteen healthy participants attended two testing sessions one week apart. Tasks included treadmill walking, treadmill running, single-leg squats, single-leg countermovement jumps, bilateral countermovement jumps, and drop vertical jumps. Keypoint data were processed using direct kinematics as well as in OpenSim using a full-body musculoskeletal model and inverse kinematics. Kinematic methods were compared using statistical parametric mapping and between-day reliability was calculated using intraclass correlation coefficients, mean absolute error, and minimal detectable change. Keypoint-derived inverse kinematics resulted in significantly smaller hip flexion (range = -9 to -2°), hip abduction (range = -3 to -2°), knee flexion (range = -5° to -2°), and greater dorsiflexion angles (range = 6-15°) than direct kinematics. Both markerless kinematic methods had high between-day reliability (inverse kinematics ICC 95 %CI = 0.83-0.90; direct kinematics ICC 95 %CI = 0.80-0.93). For certain tasks and joints, keypoint-derived inverse kinematics resulted in greater reliability (up to 0.47 ICC) and smaller minimal detectable changes (up to 13°) than direct kinematics. Performing 2-4 trials was sufficient to maximize reliability for most tasks. A single markerless depth camera can reliably measure lower limb joint angles, and skeletal model-constrained inverse kinematics improves lower limb joint angle reliability for certain tasks and joints.

Validation of OpenCap: A low-cost markerless motion capture system for lower-extremity kinematics during return-to-sport tasks

Low-cost markerless motion capture systems offer the potential for 3D measurement of joint angles during human movement. This study aimed to validate a smartphone-based markerless motion capture system's (OpenCap) derived lower extremity kinematics during common return-to-sport tasks, comparing it to an established optoelectronic motion capture system. Athletes with prior anterior cruciate ligament reconstruction (12-18 months post-surgery) performed three movements: a jump-landing-rebound, single-leg hop, and lateral-vertical hop. Kinematics were recorded concurrently with two smartphones running OpenCap's software and with a 10-camera, marker-based motion capture system. Validity of lower extremity joint kinematics was assessed across 437 recorded trials using measures of agreement (coefficient of multiple correlation: CMC) and error (mean absolute error: MAE, root mean squared error: RMSE) across the time series of movement. Agreement was best in the sagittal plane for the knee and hip in all movements (CMC > 0.94), followed by the ankle (CMC = 0.84-0.93). Lower agreement was observed for frontal (CMC = 0.47-0.78) and transverse (CMC = 0.51-0.6) plane motion. OpenCap presented a grand mean error of 3.85° (MAE) and 4.34° (RMSE) across all joint angles and movements. These results were comparable to other available markerless systems. Most notably, OpenCap's user-friendly interface, free software, and small physical footprint have the potential to extend motion analysis applications beyond conventional biomechanics labs, thus enhancing the accessibility for a diverse range of users.

Enhancing public health in developing nations through smartphone-based motor assessment

Several protocols for motor assessment have been validated for use on smartphones and could be employed by public healthcare systems to monitor motor functional losses in populations, particularly those with lower income levels. In addition to being cost-effective and widely distributed across populations of varying income levels, the use of smartphones in motor assessment offers a range of advantages that could be leveraged by governments, especially in developing and poorer countries. Some topics related to potential interventions should be considered by healthcare managers before initiating the implementation of such a digital intervention.

A Step Forward Understanding Directional Limitations in Markerless Smartphone-Based Gait Analysis: A Pilot Study

The progress in markerless technologies is providing clinicians with tools to shorten the time of assessment rapidly, but raises questions about the potential trade-off in accuracy compared to traditional marker-based systems. This study evaluated the OpenCap system against a traditional marker-based system-Vicon. Our focus was on its performance in capturing walking both toward and away from two iPhone cameras in the same setting, which allowed capturing the Timed Up and Go (TUG) test. The performance of the OpenCap system was compared to that of a standard marker-based system by comparing spatial-temporal and kinematic parameters in 10 participants. The study focused on identifying potential discrepancies in accuracy and comparing results using correlation analysis. Case examples further explored our results. The OpenCap system demonstrated good accuracy in spatial-temporal parameters but faced challenges in accurately capturing kinematic parameters, especially in the walking direction facing away from the cameras. Notably, the two walking directions observed significant differences in pelvic obliquity, hip abduction, and ankle flexion. Our findings suggest areas for improvement in markerless technologies, highlighting their potential in clinical settings.

Light-Adaptive Human Body Key Point Detection Algorithm Based on Multi-Source Information Fusion

The identification of key points in the human body is vital for sports rehabilitation, medical diagnosis, human-computer interaction, and related fields. Currently, depth cameras provide more precise depth information on these crucial points. However, human motion can lead to variations in the positions of these key points. While the Mediapipe algorithm demonstrates effective anti-shake capabilities for these points, its accuracy can be easily affected by changes in lighting conditions. To address these challenges, this study proposes an illumination-adaptive algorithm for detecting human key points through the fusion of multi-source information. By integrating key point data from the depth camera and Mediapipe, an illumination change model is established to simulate environmental lighting variations. Subsequently, the fitting function of the relationship between lighting conditions and adaptive weights is solved to achieve lighting adaptation for human key point detection. Experimental verification and similarity analysis with benchmark data yielded R2 results of 0.96 and 0.93, and cosine similarity results of 0.92 and 0.90. With a threshold range of 8, the joint accuracy rates for the two rehabilitation actions were found to be 89% and 88%. The experimental results demonstrate the stability of the proposed method in detecting key points in the human body under changing illumination conditions, its anti-shake ability for human movement, and its high detection accuracy. This method shows promise for applications in human-computer interaction, sports rehabilitation, and virtual reality.

MRI-derived Articular Cartilage Strains Predict Patient-Reported Outcomes Six Months Post Anterior Cruciate Ligament Reconstruction

Key terms

Multicontrast and Multiparametric, Magnetic Resonance Imaging, Osteoarthritis, Functional Biomechanical Imaging, Knee Joint Degeneration What is known about the subject: dualMRI has been used to quantify strains in a healthy human population in vivo and in cartilage explant models. Previously, OA severity, as determined by histology, has been positively correlated to increased shear and transverse strains in cartilage explants. What this study adds to existing knowledge: This is the first in vivo use of dualMRI in a participant demographic post-ACL reconstruction and at risk for developing osteoarthritis. This study shows that dualMRI-derived strains are more significantly correlated with patient-reported outcomes than any MRI relaxometry metric.

Background

Anterior cruciate ligament (ACL) injuries lead to an increased risk of osteoarthritis, characterized by altered cartilage tissue structure and function. Displacements under applied loading by magnetic resonance imaging (dualMRI) is a novel MRI technique that can be used to quantify mechanical strain in cartilage while undergoing a physiological load.

Purpose

To determine if strains derived by dualMRI and relaxometry measures correlate with patient-reported outcomes at six months post unilateral ACL reconstruction.

Study Design

Cohort study.

Methods

Quantitative MRI (T2, T2*, T1ρ) measurements and transverse, axial, and shear strains were quantified in the medial articular tibiofemoral cartilage of 35 participants at six-months post unilateral ACL reconstruction. The relationships between patient-reported outcomes (WOMAC, KOOS, MARS) and all qMRI relaxation times were quantified using general linear mixed-effects models. A combined best-fit multicontrast MRI model was then developed using backwards regression to determine the patient features and MRI metrics that are most predictive of patient-reported outcome scores.

Results

Higher femoral strains were significantly correlated with worse patient-reported functional outcomes. Femoral shear and transverse strains were positively correlated with six-month KOOS and WOMAC scores, after controlling for covariates. No relaxometry measures were correlated with patient-reported outcome scores. We identified the best-fit model for predicting WOMAC score using multiple MRI measures and patient-specific information, including sex, age, graft type, femoral transverse strain, femoral axial strain, and femoral shear strain. The best-fit model significantly predicted WOMAC score (p<0.001) better than any one individual MRI metric alone. When we regressed the model-predicted WOMAC scores against the patient-reported WOMAC scores, we found that our model achieved a goodness of fit exceeding 0.52.

Conclusions

This work presents the first use of dualMRI in vivo in a cohort of participants at risk for developing osteoarthritis. Our results indicate that both shear and transverse strains are highly correlated with patient-reported outcome severity could serve as novel imaging biomarkers to predict the development of osteoarthritis.

IMU-Based Real-Time Estimation of Gait Phase Using Multi-Resolution Neural Networks

This work presents a real-time gait phase estimator using thigh- and shank-mounted inertial measurement units (IMUs). A multi-rate convolutional neural network (CNN) was trained to estimate gait phase for a dataset of 16 participants walking on an instrumented treadmill with speeds varying between 0.1 to 1.9 m/s, and conditions such as asymmetric walking, stop-start, and sudden speed changes. One-subject-out cross-validation was used to assess the robustness of the estimator to the gait patterns of new individuals. The proposed model had a spatial root mean square error of 5.00±1.65%, and a temporal mean absolute error of 2.78±0.97% evaluated at the heel strike. A second cross-validation was performed to show that leaving out any of the walking conditions from the training dataset did not result in significant performance degradation. A 2-sample Kolmogorov-Smirnov test showed that there was no significant increase in spatial or temporal error when testing on the abnormal walking conditions left out of the training set. The results of the two cross-validations demonstrate that the proposed model generalizes well across new participants, various walking speeds, and gait patterns, showcasing its potential for use in investigating patient populations with pathological gaits and facilitating robot-assisted walking.

Estimating 3D kinematics and kinetics from virtual inertial sensor data through musculoskeletal movement simulations

Portable measurement systems using inertial sensors enable motion capture outside the lab, facilitating longitudinal and large-scale studies in natural environments. However, estimating 3D kinematics and kinetics from inertial data for a comprehensive biomechanical movement analysis is still challenging. Machine learning models or stepwise approaches performing Kalman filtering, inverse kinematics, and inverse dynamics can lead to inconsistencies between kinematics and kinetics. We investigated the reconstruction of 3D kinematics and kinetics of arbitrary running motions from inertial sensor data using optimal control simulations of full-body musculoskeletal models. To evaluate the feasibility of the proposed method, we used marker tracking simulations created from optical motion capture data as a reference and for computing virtual inertial data such that the desired solution was known exactly. We generated the inertial tracking simulations by formulating optimal control problems that tracked virtual acceleration and angular velocity while minimizing effort without requiring a task constraint or an initial state. To evaluate the proposed approach, we reconstructed three trials each of straight running, curved running, and a v-cut of 10 participants. We compared the estimated inertial signals and biomechanical variables of the marker and inertial tracking simulations. The inertial data was tracked closely, resulting in low mean root mean squared deviations for pelvis translation (≤20.2 mm), angles (≤1.8 deg), ground reaction forces (≤1.1 BW%), joint moments (≤0.1 BWBH%), and muscle forces (≤5.4 BW%) and high mean coefficients of multiple correlation for all biomechanical variables ( ≥ 0.99 ) . Accordingly, our results showed that optimal control simulations tracking 3D inertial data could reconstruct the kinematics and kinetics of individual trials of all running motions. The simulations led to mutually and dynamically consistent kinematics and kinetics, which allows researching causal chains, for example, to analyze anterior cruciate ligament injury prevention. Our work proved the feasibility of the approach using virtual inertial data. When using the approach in the future with measured data, the sensor location and alignment on the segment must be estimated, and soft-tissue artifacts are potential error sources. Nevertheless, we demonstrated that optimal control simulation tracking inertial data is highly promising for estimating 3D kinematics and kinetics for a comprehensive biomechanical analysis.

Multimodal human motion dataset of 3D anatomical landmarks and pose keypoints

In this paper, we present a dataset that takes 2D and 3D human pose keypoints estimated from images and relates them to the location of 3D anatomical landmarks. The dataset contains 51,051 poses obtained from 71 persons in A-Pose while performing 7 movements (walking, running, squatting, and four types of jumping). These poses were scanned to build a collection of 3D moving textured meshes with anatomical correspondence. Each mesh in that collection was used to obtain the 3D locations of 53 anatomical landmarks, and 48 images were created using virtual cameras with different perspectives. 2D pose keypoints from those images were obtained using the MediaPipe Human Pose Landmarker, and their corresponding 3D keypoints were calculated by linear triangulation. The dataset consists of a folder for each participant containing two Track Row Column (TRC) files and one JSON file for each movement sequence. One TRC file is used to store the 3D data of the triangulated 3D keypoints while the other contains the 3D anatomical landmarks. The JSON file is used to store the 2D keypoints and the calibration parameters of the virtual cameras. The anthropometric characteristics of the participants are annotated in a single CSV file. These data are intended to be used in developments that require the transformation of existing human pose solutions in computer vision into biomechanical applications or simulations. This dataset can also be used in other applications related to training neural networks for human motion analysis and studying their influence on anthropometric characteristics.

Exercise quantification from single camera view markerless 3D pose estimation

Sports physiotherapists and coaches are tasked with evaluating the movement quality of athletes across the spectrum of ability and experience. However, the accuracy of visual observation is low and existing technology outside of expensive lab-based solutions has limited adoption, leading to an unmet need for an efficient and accurate means to measure static and dynamic joint angles during movement, converted to movement metrics useable by practitioners. This paper proposes a set of pose landmarks for computing frequently used joint angles as metrics of interest to sports physiotherapists and coaches in assessing common strength-building human exercise movements. It then proposes a set of rules for computing these metrics for a range of common exercises (single and double drop jumps and counter-movement jumps, deadlifts and various squats) from anatomical key-points detected using video, and evaluates the accuracy of these using a published 3D human pose model trained with ground truth data derived from VICON motion capture of common rehabilitation exercises. Results show a set of mathematically defined metrics which are derived from the chosen pose landmarks, and which are sufficient to compute the metrics for each of the exercises under consideration. Comparison to ground truth data showed that root mean square angle errors were within 10° for all exercises for the following metrics: shin angle, knee varus/valgus and left/right flexion, hip flexion and pelvic tilt, trunk angle, spinal flexion lower/upper/mid and rib flare. Larger errors (though still all within 15°) were observed for shoulder flexion and ASIS asymmetry in some exercises, notably front squats and drop-jumps. In conclusion, the contribution of this paper is that a set of sufficient key-points and associated metrics for exercise assessment from 3D human pose have been uniquely defined. Further, we found generally very good accuracy of the Strided Transformer 3D pose model in predicting these metrics for the chosen set of exercises from a single mobile device camera, when trained on a suitable set of functional exercises recorded using a VICON motion capture system. Future assessment of generalization is needed.

Hamstrings are stretched more and faster during accelerative running compared to speed-matched constant speed running

Background

Hamstring strain injuries are associated with significant time away from sport and high reinjury rates. Recent evidence suggests that hamstring injuries often occur during accelerative running, but investigations of hamstring mechanics have primarily examined constant speed running on a treadmill. To help fill this gap in knowledge, this study compares hamstring lengths and lengthening velocities between accelerative running and constant speed overground running.

Methods

We recorded 2 synchronized videos of 10 participants (5 female, 5 male) during 6 accelerative running trials and 6 constant speed running trials. We used OpenCap (a markerless motion capture system) to estimate body segment kinematics for each trial and a 3-dimensional musculoskeletal model to compute peak length and step-average lengthening velocity of the biceps femoris (long head) muscle-tendon unit. To compare running conditions, we used linear mixed regression models with running speed (normalized by the subject-specific maximum) as the independent variable.

Results

At running speeds below 75% of top speed accelerative running resulted in greater peak lengths than constant speed running. For example, the peak hamstring muscle-tendon length when a person accelerated from running at only 50% of top speed was equivalent to running at a constant 88% of top speed. Lengthening velocities were greater during accelerative running at all running speeds. Differences in hip flexion kinematics primarily drove the greater peak muscle-tendon lengths and lengthening velocities observed in accelerative running.

Conclusion

Hamstrings are subjected to longer muscle-tendon lengths and faster lengthening velocities in accelerative running compared to constant speed running. This provides a biomechanical explanation for the observation that hamstring strain injuries often occur during acceleration. Our results suggest coaches who monitor exposure to high-risk circumstances (long lengths, fast lengthening velocities) should consider the accelerative nature of running in addition to running speed.

Clinical gait analysis using video-based pose estimation: Multiple perspectives, clinical populations, and measuring change

Gait dysfunction is common in many clinical populations and often has a profound and deleterious impact on independence and quality of life. Gait analysis is a foundational component of rehabilitation because it is critical to identify and understand the specific deficits that should be targeted prior to the initiation of treatment. Unfortunately, current state-of-the-art approaches to gait analysis (e.g., marker-based motion capture systems, instrumented gait mats) are largely inaccessible due to prohibitive costs of time, money, and effort required to perform the assessments. Here, we demonstrate the ability to perform quantitative gait analyses in multiple clinical populations using only simple videos recorded using low-cost devices (tablets). We report four primary advances: 1) a novel, versatile workflow that leverages an open-source human pose estimation algorithm (OpenPose) to perform gait analyses using videos recorded from multiple different perspectives (e.g., frontal, sagittal), 2) validation of this workflow in three different populations of participants (adults without gait impairment, persons post-stroke, and persons with Parkinson's disease) via comparison to ground-truth three-dimensional motion capture, 3) demonstration of the ability to capture clinically relevant, condition-specific gait parameters, and 4) tracking of within-participant changes in gait, as is required to measure progress in rehabilitation and recovery. Importantly, our workflow has been made freely available and does not require prior gait analysis expertise. The ability to perform quantitative gait analyses in nearly any setting using only low-cost devices and computer vision offers significant potential for dramatic improvement in the accessibility of clinical gait analysis across different patient populations.

Inter-trial variability is higher in 3D markerless compared to marker-based motion capture: Implications for data post-processing and analysis

Markerless motion capture has recently attracted significant interest in clinical gait analysis and human movement science. Its ease of use and potential to streamline motion capture recordings bear great potential for out-of-the-laboratory measurements in large cohorts. While previous studies have shown that markerless systems can achieve acceptable accuracy and reliability for kinematic parameters of gait, they also noted higher inter-trial variability of markerless data. Since increased inter-trial variability can have important implications for data post-processing and analysis, this study compared the inter-trial variability of simultaneously recorded markerless and marker-based data. For this purpose, the data of 18 healthy volunteers were used who were instructed to simulate four different gait patterns: physiological, crouch, circumduction, and equinus gait. Gait analysis was performed using the smartphone-based markerless system OpenCap and a marker-based motion capture system. We compared the inter-trial variability of both systems and also evaluated if changes in inter-trial variability may depend on the analyzed gait pattern. Compared to the marker-based data, we observed an increase of inter-trial variability for the markerless system ranging from 6.6% to 22.0% for the different gait patterns. Our findings demonstrate that the markerless pose estimation pipelines can introduce additionally variability in the kinematic data across different gait patterns and levels of natural variability. We recommend using averaged waveforms rather than single ones to mitigate this problem. Further, caution is advised when using variability-based metrics in gait and human movement analysis based on markerless data as increased inter-trial variability can lead to misleading results.

Jumping towards field-based ground reaction force estimation and assessment with OpenCap

Low-cost and field-viable methods that can simultaneously assess external kinetics and kinematics are necessary to enhance field-based biomechanical monitoring. The aim of this study was to determine the accuracy and usability of ground reaction force (GRF) profiles estimated from segmental kinematics, measured with OpenCap (a low-cost markerless motion-capture system), during common jumping movements. Full-body segmental kinematics were recorded for fifteen recreational athletes performing countermovement, squat, bilateral drop, and unilateral drop jumps, and used to estimate vertical GRFs with a mechanics-based method. Eleven distinct performance-, fatigue-, or injury-related GRF variables were then validated against a gold-standard force platform. Across jumping movements, a total of six and three GRF variables were estimated with a bias or limits of agreement <5 % respectively. Bias and limits of agreement were between 5 and 15 % for seventeen and nineteen variables respectively. Moreover, we show that estimated force variables with a bias <15 % can adequately assess the within-athlete changes in GRF variables between jumping conditions (arm swing or leg dominance). These findings indicate that using a low-cost and field-viable markerless motion capture system (OpenCap) to estimate and assess GRF profiles during common jumping movements is approaching acceptable limits of accuracy. The presented method can be used to monitor force variables of interest and examine underlying segmental kinematics. This application is a jump towards researchers and sports practitioners performing biomechanical monitoring of jumping efficiently, regularly, and extensively in field settings.