Markerless motion capturing has the potential to provide a low-cost and accessible alternative to traditional marker-based systems for real-world biomechanical assessment. However, before these systems can be put into practice, we need to rigorously evaluate their accuracy in estimating joint kinematics for various gait patterns. This study evaluated the accuracy of a low-cost, open-source, and smartphone-based markerless motion capture system, namely OpenCap, for measuring 3D joint kinematics in healthy and pathological gait compared to a marker-based system. 21 healthy volunteers were instructed to walk with four different gait patterns: physiological, crouch, circumduction, and equinus gait. Three-dimensional kinematic data were simultaneously recorded using the markerless and a marker-based motion capture system. The root mean square error (RMSE) and the peak error were calculated between every joint kinematic variable obtained by both systems. We found an overall RMSE of 5.8 (SD: 1.8 degrees) and a peak error of 11.3 degrees (SD: 3.9). A repeated measures ANOVA with post hoc tests indicated significant differences in RMSE and peak errors between the four gait patterns (p ¡ 0.05). Physiological gait presented the lowest, crouch and circumduction gait the highest errors. Our findings indicate a roughly comparable accuracy to IMU-based approaches and commercial markerless multi-camera solutions. However, errors are still above clinically desirable thresholds of two to five degrees. While our findings highlight the potential of markerless systems for assessing gait kinematics, they also underpin the need to further improve the underlying deep learning algorithms to make markerless pose estimation a valuable tool in clinical settings.

The quantification of ground reaction forces (GRF) is a standard tool for clinicians to quantify and analyze human locomotion. Such recordings produce a vast amount of complex data and variables which are difficult to comprehend. This makes data interpretation challenging. Machine learning approaches seem to be promising tools to support clinicians in identifying and categorizing specific gait patterns. However, the quality of such approaches strongly depends on the amount of available annotated data to train the underlying models. Therefore, we present GaitRec, a comprehensive and completely annotated large-scale dataset containing bi-lateral GRF walking trials of 2,084 patients with various musculoskeletal impairments and data from 211 healthy controls. The dataset comprises data of patients after joint replacement, fractures, ligament ruptures, and related disorders at the hip, knee, ankle or calcaneus during their entire stay(s) at a rehabilitation center. The data sum up to a total of 75,732 bi-lateral walking trials and enable researchers to classify gait patterns at a large-scale as well as to analyze the entire recovery process of patients.

This study investigates the accuracy of the tracking system LPM (Local Position Measurement). The goal was to determine detailed error values of the system in the context of sports performance analyses. Six moderately trained male soccer players (amateur level) performed 276 runs on three different courses at six different speeds. Additionally, ten small-sided game plays were carried out. All runs and game plays were recorded with the LPM tracking system and the motion capture system VICON simultaneously. VICON served as the reference system. The absolute error of all LPM position estimations was on average 23.4±20.7 cm. The estimation for average velocities varied between 0.01 km h⁻¹ and 0.23 km h⁻¹, the maximum speed estimations differed by up to 2.71 km h⁻¹. In addition, the results showed that the accuracy of the LPM system is highly dependent on the instantaneous dynamics of the player and decreases in the margins of the observation field. These dependencies were quantified. Considering commonly used applications of position tracking systems in sports (Leser, Ogris, & Baca, 2011 Leser, R. Analyses of tacklings in football. Presented at the 8th International Symposium of the International Association of Computer Science in Sports, Shanghai, China. 22 September [Google Scholar] ), the accuracy of LPM is acceptable for position and velocity estimations. The system provides valuable results for average velocities but seems to be far less reliable when dealing with high dynamic movements and measuring instantaneous velocities.

This article proposes a comprehensive investigation of the automatic classification of functional gait disorders based solely on ground reaction force (GRF) measurements. The aim of the study is twofold: (1) to investigate the suitability of stateof-the-art GRF parameterization techniques (representations) for the discrimination of functional gait disorders; and (2) to provide a first performance baseline for the automated classification of functional gait disorders for a large-scale dataset. The utilized database comprises GRF measurements from 279 patients with gait disorders (GDs) and data from 161 healthy controls (N). Patients were manually classified into four classes with different functional impairments associated with the “hip”, “knee”, “ankle”, and “calcaneus”. Different parameterizations are investigated: GRF parameters, global principal component analysis (PCA)-based representations and a combined representation applying PCA on GRF parameters. The discriminative power of each parameterization for different classes is investigated by linear discriminant analysis (LDA). Based on this analysis, two classification experiments are pursued: (1) distinction between healthy and impaired gait (N vs. GD) and (2) multi-class classification between healthy gait and all four GD classes. Experiments show promising results and reveal among others that several factors, such as imbalanced class cardinalities and varying numbers of measurement sessions per patient have a strong impact on the classification accuracy and therefore need to be taken into account. The results represent a promising first step towards the automated classification of gait disorders and a first performance baseline for future developments in this direction.

In 2014, more than 10 million people in the US were affected by an ambulatory disability. Thus, gait rehabilitation is a crucial part of health care systems. The quantification of human locomotion enables clinicians to describe and analyze a patient's gait performance in detail and allows them to base clinical decisions on objective data. These assessments generate a vast amount of complex data which need to be interpreted in a short time period. We conducted a design study in cooperation with gait analysis experts to develop a novel Knowledge-Assisted Visual Analytics solution for clinical Gait analysis (KAVAGait). KAVAGait allows the clinician to store and inspect complex data derived during clinical gait analysis. The system incorporates innovative and interactive visual interface concepts, which were developed based on the needs of clinicians. Additionally, an explicit knowledge store (EKS) allows externalization and storage of implicit knowledge from clinicians. It makes this information available for others, supporting the process of data inspection and clinical decision making. We validated our system by conducting expert reviews, a user study, and a case study. Results suggest that KAVAGait is able to support a clinician during clinical practice by visualizing complex gait data and providing knowledge of other clinicians.

Virtual reality (VR) is an emerging technology offering tremendous opportunities to aid gait rehabilitation. To this date, real walking with users immersed in virtual environments with head-mounted displays (HMDs) is either possible with treadmills or room-scale (overground) VR setups. Especially for the latter, there is a growing interest in applications for interactive gait training as they could allow for more self-paced and natural walking. This study investigated if walking in an overground VR environment has relevant effects on 3D gait biomechanics. A convenience sample of 21 healthy individuals underwent standard 3D gait analysis during four randomly assigned walking conditions: the real laboratory (RLab), a virtual laboratory resembling the real world (VRLab), a small version of the VRlab (VRLab−), and a version which is twice as long as the VRlab (VRLab+). To immerse the participants in the virtual environment we used a VR-HMD, which was operated wireless and calibrated in a way that the virtual labs would match the real-world. Walking speed and a single measure of gait kinematic variability (GaitSD) served as primary outcomes next to standard spatio-temporal parameters, their coefficients of variant (CV%), kinematics, and kinetics. Briefly described, participants demonstrated a slower walking pattern (−0.09 ± 0.06 m/s) and small accompanying kinematic and kinetic changes. Participants also showed a markedly increased gait variability in lower extremity gait kinematics and spatio-temporal parameters. No differences were found between walking in VRLab+ vs. VRLab−. Most of the kinematic and kinetic differences were too small to be regarded as relevant, but increased kinematic variability (+57%) along with increased percent double support time (+4%), and increased step width variability (+38%) indicate gait adaptions toward a more conservative or cautious gait due to instability induced by the VR environment. We suggest considering these effects in the design of VR-based overground training devices. Our study lays the foundation for upcoming developments in the field of VR-assisted gait rehabilitation as it describes how VR in overground walking scenarios impacts our gait pattern. This information is of high relevance when one wants to develop purposeful rehabilitation tools.