Computational musculoskeletal models estimate the internal forces that produce human movement by representing the bones, muscles, and joints of the body. The accuracy of these simulations depends on how well the model reflects the anatomy of the individual. Current practice personalises models by linearly scaling a generic template. Scaling cannot accurately reproduce internal anatomical landmarks such as hip joint centres, resulting in deviations of 20-40mm from imaging-based ground truth.
This study develops a method to predict anatomical landmark positions of the lower limbs directly from external skin surface geometry. The proposed approach achieves a mean prediction error of 25.2mm across 18 lower-limb landmarks. These are of the same order as the hip joint centre deviations reported for linear scaling.
The results show that external surface geometry contains sufficient information to estimate these internal anatomical locations, and that the regression model captures this relationship across participants. Prediction accuracy is limited by the consistency of the input skin representation.
These findings show that anatomical landmarks relevant for musculoskeletal modelling can be estimated from skin geometry, providing a non-invasive approach to obtaining subject-specific anatomical information. This establishes a framework for exploring surface-based methods for musculoskeletal model personalisation. ... Read more

Background and Objective: Human walking is governed by the interaction between central pattern generators (CPGs), reflexes, and the vestibular system. While single-layer CPG models like the Unit Burst Generator (UBG) are commonly used in simulations, they couple oscillation frequency and amplitude, limiting control of gait velocity. This study evaluates a two-layer UBG (TLU) architecture, which decouples these properties, to determine its efficacy in producing realistic human gait and its ability to control gait velocity.

Method: A 2D musculoskeletal model (9 DOF, 18 muscles) was controlled by the UBG or TLU architecture during gait, integrated with muscle reflexes and vestibular feedback. Additionally, velocity control was assessed by optimizing specific CPG parameters to increase walking speed.

Results: Both controllers produced stable, physiologically plausible gait that aligned well with normative data. Neural analysis showed that while reflexes mainly controlled lower leg muscles, the CPG was essential for the loading response in muscles spanning the knee and hip joints. In the velocity control, the TLU model could reach higher gait velocities with fewer optimization parameters compared to the UBG.

Conclusion: The CPG effectively coordinates with muscle reflexes and vestibular feedback to produce human-like gait, complementing where the other neural control mechanisms fall short. The TLU provides a more efficient mechanism for gait modulation than the UBG by separating frequency and amplitude control. ... Read more

Research on 3D predictive gait simulation remains limited. This study therefore verifies an existing 3D predictive gait simulation framework implemented in SCONE, with the goal that this verified framework can serve as a physiologically plausible modelling and optimisation platform for future related studies. Four optimisation criteria were selected, namely minimising cost of transport, minimising muscle activity, maxmising head instability, and minimising foot-ground impact. These criteria were combined to form a set of objective functions, under which the framework was optimised. The predicted results produced by each objective function were quantitatively compared with experimental data using Pearson r and RMSE/SD, and agreement was evaluated across multiple biomechanical categories, including joint kinematics, ground reaction forces, joint moments and joint powers. The results indicate that, under the optimal objective function, the predictive performance approaches that of the ExpTrack solution that directly tracks experimental data. The framework reproduces sagittal plane hip, knee and ankle angles, vertical and anterior-posterior ground reaction forces, all joint moments and ankle power with strong agreement, but agreement is weaker for hip adduction, medio-lateral ground reaction force, and knee and hip power. Overall, these findings demonstrate the strengths of the framework in reproducing experimental 3D gait, while also revealing limitations in medio-lateral stability control and in the current model and controller settings, providing a basis for targeted improvements in future work. ... Read more

Background: Anterior cruciate ligament (ACL) injuries commonly reduce knee stability and increase joint loading, often leading to compensatory gait alterations that may increase injury risk. Functional electrical stimulation (FES) of the biceps femoris long head (BFLH) during the gait stance may improve knee stability by reducing harmful joint loading, but the effects on voluntary muscle control remain unclear.

Research question: This study examined whether FES of the BFLH during the stance phase of the gait reduces ACL-relevant knee joint loading in healthy adults and whether it alters voluntary muscle control. Additionally, the use of gluteus maximus (GLMAX) sEMG as a proxy for BFLH activation was assessed.

Method: Nine healthy participants walked on a treadmill under control and FES-assisted conditions. Kinematic, kinetic, and sEMG data were analyzed using statistical parametric mapping and linear mixed-effects models.

Results: FES of the BFLH significantly reduced internal knee rotation moment (KRM) with 9.37% during 42–48% of the gait cycle (p = 0.0002; d = 0.42). Knee adduction moment (KAM) showed non-significant reductions in both legs (non-stimulated: p = 0.0317, d = 0.18; stimulated: p = 0.0492, d = 0.37). Knee abduction angle (KAA) and knee rotation angle (KRA) showed no significant changes (p > 0.05). In sEMG analysis, inconsistent timing between GLMAX and BFLH activation indicated GLMAX is not a reliable surrogate for estimating BFLH activity. Regarding voluntary control, only peak KAM increased slightly over strides during FES-assisted walking (p = 0.006), possibly due to muscle fatigue. No significant retention or after-effects were observed.

Conclusion: Targeted FES of the BFLH can reduce ACL-relevant knee loading without impairing voluntary motor control. sEMG results highlight the need for direct BFLH monitoring, as GLMAX is an unreliable proxy. These findings support further exploration of FES strategies for ACL injury prevention and rehabilitation.
... Read more

Abstract - Background Most widely used musculoskeletal models are predominantly based on male anatomy. This limits accurate biomechanical analysis in women, despite notable sex differences in occurrence of musculoskeletal pathologies. Aim This study aimed to develop a female musculoskeletal model of the lower extremity (YONI). The YONI is compared to the generic (male-based) model in simulation with a female participant. Methods The YONI model was developed in OpenSim Creator based on MR images. Comparisons were done between the YONI model and a scaled RAJAG model and a personalized model of using motion capture data in simulations. Results For comparison between YONI and RAJAG with another female subject, mean RMS error for the personalized model was 0.0096 m (SD = 0.0013), for the scaled RAJAG model 0.0247 m (SD = 0.0268) and for the scaled YONI model 0.0097 m (SD = 0.0010). Although SPM paired t-test showed significant differences for both YONI and RAJAG compared to the personalized model for all joint angles, the YONI model showed lower t-values compared to RAJAG. Joint moments reveal larger differences between YONI and scaled RAJAG models in the hip angles during the swing phase. Reserve moments were low in hip flexion and hip adduction, but higher in knee flexion and ankle flexion. Conclusion While observed kinematic and dynamic differences require cautious interpretation due to model limitations and data constraints, this work represents a crucial step toward the development of a female musculoskeletal model, essential for advancing biomechanical research and clinical applications for women. ... Read more

Despite the recognized impact of sex on biomechanics, research remains biased toward male anatomy, raising concerns about the validity of musculoskeletal (MSK) model predictions for females. This study investigated whether sex-specific bone geometry variations predict differences in the proportional volumes of the Gluteus Maximus (GMAX) and Rectus Femoris (RFEM). Using an MRI-based nnU-Net segmentation model trained in this thesis, bone metrics were extracted, and muscle volumes were normalized to derive proportional volumes for 16 young adults (9F/7M). The segmentation model demonstrated high accuracy (DSC: 0.926 for bones, 0.954 for muscles), revealing significant sexual dimorphism in bone geometry. Males exhibited greater femoral offsets and knee widths, while females had larger posterior pelvic widths and depths. %RFEM was significantly higher in males (p = 0.01), but %GMAX showed no sex-related differences. Regression analysis identified femoral offset and femur length as partial predictors of %RFEM (R^2 = 0.478), with pelvis-femur length weakly predicting %GMAX (R^2 = 0.151). However, the low predictive power suggests limitations in using bone metrics to estimate muscle volume proportions. These findings indicate that femoral dimorphism may partially explain sex-related %RFEM differences, but its role in %GMAX remains unclear. Future research integrating additional biomechanical factors could enhance sex-specific MSK modeling accuracy. ... Read more

The optimal technique for individual speed skaters remains poorly understood, due to the complex interplay of technique variables (like stroke frequency, skate trajectory and push-off mechanics). Optimization with a biomechanical model can help to identify the most efficient techniques for individual skaters. This research aimed to use a validated model of a speed skater (van der Kruk et al. 2017) within an optimization framework to investigate how the optimal speed skating techniques on the straightaways are influenced by individual characteristics and environmental conditions.

Finding the optimal technique that either minimizes effort at a target velocity or maximizes velocity, was formulated as an optimal control problem and solved using direct collocation. Across different optimizations, stroke frequency, mass, leg length, air and ice friction and limits on average and maximal power were incrementally varied.

Variations in velocity and stroke frequency most clearly influenced the optimal technique. Conditions requiring less energy (optimizations for low velocity, low ice or air friction), optimized towards energy-efficient strokes with longer gliding phases and minimal lateral forces. Conditions with higher speeds and frequencies converged to longer, forceful push-offs. These push-offs maximized leg extension by descending into a deep crouched position to emphasize a sideways push-off. Generally, optimized techniques adapted a small steer angle during the gliding phase to prioritize forward gain, and larger steering angles during the push-off to direct push-off forces forward. Optimizations for higher frequencies adopted more narrow strokes, and reached higher maximized speeds. Regarding personal characteristics, increasing the model's average and maximal power limits most significantly increased maximal velocity. ... Read more

Musculoskeletal modelling forms an important asset in research on understanding human motions and neuromuscular performance. The reliability of musculoskeletal model outcomes depend on the accuracy of the parameters, particularly Optimal Fiber Length (OFL). However, the OFL may vary considerably between populations. To date it is unknown how such variations might affect musculoskeletal modelling. Therefore, in vivo determination of OFL is essential for subject-specific models, and understanding population-specific differences for improvement of model diversity. This study developed an in vivo methodology for measuring the OFL that is easily accessible for large scale implementations. It was applied it to the m. rectus femoris for validation.

OFL was derived from the force-length relationship by measuring muscle force, calculated from knee moments, muscle-tendon moment arm, and fascicle length using ultrasound. Muscle activation was standardized via electrical stimulation. The protocol was separately evaluated for validity, reliability, and usability.

Results indicated that knee moment and muscle-tendon moment arm measurements deviated from literature values due to experimental setup limitation, and active fascicle lengths could not be reliably estimated due to the complex muscle architecture. Consequently, the current approach did not yield valid OFL estimates. This study provides insight into the challenges of developing reliable in vivo measurement techniques.

Future studies should employ improved experimental approaches such as increasing electrical muscle stimulation, use of dynamometry and more advanced ultrasound techniques, and applications to other muscles and joints, ultimately, providing a foundation for in vivo estimation of OFL, facilitating investigation of population-specific differences and improving diversity in musculoskeletal modelling. ... Read more

The metabolic cost of walking reflects gait efficiency and is influenced by biomechanical factors as walking speed. While most walking research uses treadmills, older adults are found to exhibit a greater elevation in the cost of walking on the treadmill compared to overground walking than younger adults. A possible cause for this elevation could be the higher stability demands in older adults during treadmill walking, possibly influenced by a higher vestibular contribution. This study investigated whether increased vestibular demands for balance control contribute to this elevated cost in older adults. Ten younger (mean age 26.4 years) and ten older adults (mean age 68.6 years) completed 5-minute treadmill and overground walking trials at preferred and slow fixed speeds. Metabolic cost was measured, and vestibular contributions to balance were assessed via electrical vestibular stimulation, which induced virtual movements and evoked balance correcting responses measured by inertial sensors on the back and ankles. Treadmill walking increased the cost of walking significantly by 15-23% compared to overground walking, with no significant age effect. Vestibular stimulation increased metabolic cost significantly in both overground and treadmill walking and age groups. Assessment of the vestibular contributions to kinematic measures revealed a significant increase in vestibular contribution to balance at slower walking speeds, but no significant effect of age and no large effect of treadmill or overground. Indicating that the measured participants cannot conclude that the elevation of cost in treadmill walking in older adults is due to the vestibular contribution to balance. ... Read more

Age-related decline in physical and neural capacity can make the sit-to-stand (STS) motion increasingly difficult for older adults, significantly impacting their quality of life. Despite these declines, humans adopt compensatory movement strategies to mitigate the effects of reduced capacity, maintaining functional mobility. Predictive simulations offer a tool for studying the relationship between capacity decline and compensation strategies. However, previous predictive studies have omitted the modeling and control of arm movements, thus neglecting key arm compensation strategies relevant to the STS motion. Therefore, this study aimed to develop a neuromuscular controller for a three-dimensional musculoskeletal model that includes the arms, enabling the simulation of STS arm compensation strategies. STS arm strategies were successfully simulated and displayed comparable joint kinematics with experimental data. However, the simulations revealed elevated leg muscle activations and an overestimated vertical ground reaction force. Additional simulations with changed conditions demonstrated the effective use of the armrest and thigh push-off strategies to adapt to lower seat heights and reduce peak knee joint load. Overall, the neuromuscular controller in this study provides a new basis for future STS research into uncovering the link between capacity decline and compensation strategies, potentially leading to improved methods for assessing and addressing age-related declines in crucial movements. ... Read more

Transfemoral prosthesis users typically perform the sit-to-stand motion unilaterally, placing minimal load on the prosthetic side. This increases the risk of injury and accelerates the degeneration of the intact limb. To address this, understanding compensation strategies is essential. Predictive neuromusculoskeletal modeling offers a method to investigate this. The aim of this study was to develop and validate a neuromusculoskeletal model with reflex-based muscle control to simulate the sit-tostand motion in non-amputees and transfemoral amputees with a passive prosthesis. We developed a three-dimensional sit-tostand musculoskeletal model of a non-amputee (20 degrees of freedom, 24 muscles) and a transfemoral amputee (20 degrees of freedom, 19 muscles), both incorporating a two-phase stand-up controller based on an existing two-dimensional reflex controller. We compared the simulation framework to measured kinematics, muscle activations, ground reaction forces, and existing literature on degrees of asymmetry and muscle forces. The developed framework was used to optimize prosthetic knee and ankle stiffness and damping for the sit-to-stand motion with a passive prosthesis. The simulated kinematics of the non-amputee matched measured kinematics. The prosthesis model indicated compensation strategies involving increased thoracic and lumbar extension, lumbar bending towards the non-amputated side,
increased pelvic tilt combined with decreased hip flexion, and heightened muscle activation and force. Optimization results suggested a knee stiffness of 0.1432 [Nm/deg] and damping of 0.0246 [Nm·s/deg], while the ankle required a stiffness of 0.1968 [Nm/deg] and damping of 0.1350 [Nm·s/deg]. These values are recommended for testing in future experiments. ... Read more

The aim of this Master’s thesis is to determine how user engagement and motivation can be improved when using the Athlete Management System (AMS) as a monitoring system. This system monitors athletes subjectively and objectively for short and long-term progress. Increased use of this monitoring system will result in a more complete and valuable data set for coaches to identify the risks of under- or overtraining for KNSB talented athletes in both long and shorttrack disciplines.

Several interviews were conducted with athletes, experts from the KNSB Talent Teams (KTTs), and other experts in the field of AMS, sports psychology,
and sports innovation centers. These interviews were used to explore how the target group reflected on their sports’ progression and how the feedback process from the KTT staff played a role in this, supported by a literature review of related topics in the context of the project aim.

After interviews with athletes, coaches, and embedded scientists, potential design directions were identified and one was selected by evaluating the directions for feasibility and impact concerning the project aim. The chosen design opportunity is to improve communication about AMS between athletes and their coaches and to provide more guidance in an athlete’s reflective process when they need to measure their recovery-stress state of ‘Mental Readiness’ in AMS, which can be described as the athlete’s ability to concentrate on the execution of a training session. Measuring the recovery-stress state of an athlete can help identify the risk of under- or overtraining. Athletes and coaches experience difficulties interpreting and assessing ‘Mental Readiness’.

Brainstorming and concept validation sessions are conducted to develop a final design: a workshop session consisting of a presentation with three assignments to allow athletes and their coach to share their interpretations of the ‘Mental Readiness’ scale and to give first steps of guidance on how to reflect as an athlete on this scale. The final design is an addition to the kick-off meeting at the beginning of the speed skating season. It also proposes a roadmap for the long-term implementation of the final design in the context of the target group, including additional suggestions for other workshops and presentations to improve communication and behavior around AMS based on the insights from the interviews.

Further research should investigate how the final design leads to behavioral changes in user engagement and motivation in long-term implementation.
In addition, other aspects of the recovery-stress state, such as ‘emotional state’ and ‘motivation’, could be explored to broaden the communication and enhance the self-reflection of the athlete. In future research, it is important to expand guidance for athletes in their reflective capacity and for KTT staff in the correct interpretation and next steps when receiving ‘Mental Readiness’ data. This will help to motivate athletes to work with the final design and increase user engagement with the Athlete Management System. ... Read more

Achilles tendinopathy (AT) is a common Achilles tendon injury, yet its exact cause and the factors influencing progression of individuals remain unclear. Strain distribution is indicated to play a significant role in the progression, possibly linked to the twist of the subtendons. Pizzolato et al. (2020) proposed an integrated framework for studying Achilles tendon mechanics, including a finite element (FE) model estimating local displacements in the Achilles tendon. However, before im- plementing this in AT research, further testing and validation is necessary. Therefore, to make a start for future improvements, this study aims to build a foundational FE model of the Achilles tendon, verify it with in vivo local displacement data and assess the sensitivity to subtendon twist. 3D ultrasound and X-rays of the ankle provided the geometry and the moment arm of the Achilles tendon, respectively. By minimizing the error between the tendon’s elongation during contraction and the FE model prediction, the material properties were optimized. Local displacements in the sagittal and coronal plane were computed using estimated forces from in vivo studies. Simulated subtendon twists (11◦, 37◦, 65◦) examined the effect of the amount of twist on the displacement. Comparing the FE estimated local displacements to in vivo data indicated that additional substructure details are needed to accurately calculate the displacement behavior. Modifying fiber twist angles altered the uniformity of the displacements in the FE model. Therefore, further development of the FE model of the Achilles tendon is recommended before incorporating it into an Achilles tendon mechanics study. ... Read more

This study investigates different strategies that the central nervous system might adopt to solve the motor redundancy problem in young and older adults, focusing on metabolic costs, head accelerations, and gait adjustments during overground walking at various speeds. The study addresses gaps in previous research, which primarily focused on younger participants and treadmill-based trials, potentially overlooking natural gait patterns. Ten younger adults (aged 23–28) and five older adults (aged 69–77) completed eight overground walking trials at different speeds, including their preferred walking speed (PWS), predetermined speeds constant for each subject, and variations of their preferred walking speed.

Results showed that younger and older adults had similar preferred walking speeds and comparable metabolic costs when walking at their chosen pace, while younger adults exhibited higher metabolic costs at higher speeds. The PWS did not minimize the metabolic cost for either age group. At their PWS, younger adults both reduced head accelerations and maximized stability, whereas older adults prioritized stability over movement smoothness. Both groups primarily adjusted step frequency rather than step length to accommodate changes in walking speed. Additionally, no significant differences were found in maximum arm swing velocity between the two groups. These findings challenge previous assumptions about age-related differences in walking efficiency and suggest that stability may play a more critical role in gait optimization for older adults. Further research is necessary to uncover the mechanisms driving these adaptations and their impact on gait across the lifespan. ... Read more

Accurate kinematic analysis in speed skating is crucial to understand and improve skaters’ unique technique during training. Few research methods have captured joint kinematics in the past, using Inertial Measurement Units (IMUs) or manual annotation of filmed data or marker-based motion capture methods. The large motion capture volume and ice-rink environment hinders these methods to be actively adopted on rink. Thus, with growing accuracy in kinematic estimation algorithms, demand for a biomechanically accurate dataset is high. In this research, we aim to generate a virtual video dataset called ODAH-SpeedSkater, from experimental motion capture data. First, we examine the impact and accuracy of markers during motion capture. After selection of accurate inverse kinematics data, we use the SMPL-X human body model to achieve individual skater body shape and pose throughout the motion, rigging it to a skater-specific scaled OpenSim skeletal model. We render the finalized mesh sequences in an ice rink scene with skater outfits through realistic camera set-ups and configurations. Thus, we successfully create a dataset of 1,326 biomechanically annotated virtual videos of speed skating. Finally, we test our dataset on a pre-trained 3D kinematic estimation algorithm to evaluate its performance on speed skating data. In spite of limited testing, we conclude that training a network exclusively on our dataset may improve its performance, with the ultimate goal of actively implementing such networks in the rink. ... Read more

Running is one of the most practiced sports worldwide, offering numerous health benefits, but also carrying a risk of injury, mainly at the knee and ankle joints. The origin of running injuries is not fully understood. With predictive neuromusculoskeletal simulations, more insight could be gained into the biomechanical mechanisms that may lead to injuries.
However, in predictive simulations of gait, hyperextension of the knee during stance phase is often encountered. This limits their applicability in research into running-related injuries. It is unclear what causes these unrealistic kinematics, with various studies coming to conflicting conclusions.
This study aims to identify the cause of knee hyperextension in predictive models of running and subsequently, to determine the essential modeling elements for accurately simulating stance knee flexion.

A structured analysis was conducted to investigate the potential impact of the model components within the predictive simulation framework. This framework was divided into four main categories: the objective function, the musculoskeletal (MSK) model, the foot contact model, and the controller. The analysis resulted in numerous hypotheses regarding the element that might be responsible for the simulation of realistic knee kinematics. SCONE, an open-source package for neuromusculoskeletal predictive simulation, was used to test the effect of each hypothesis on the simulated running kinematics. The simulation outcomes were compared to experimental data to assess possible improvements.

The results demonstrate that, in contrast to previous literature, adaptations to the objective function, the MSK model, and the foot contact model have negligible effects on predicted running kinematics. This leads to the conclusion that the controller is essential to focus on when improving knee kinematics. Due to time constraints, multiphase control could not be implemented. Therefore, the exact reflex pathways and phase transitions should be further investigated for the predictive simulation of running before implementation is possible. ... Read more

In biomechanics, human movement studies are carried out to assess the subject’s kinematic and kinetic variables for a healthy gait. Currently, marker-based systems are the standardized method to extract the kinematic variables of subjects. The marker-based systems pose some serious challenges like cost and portability, and the calibration and synchronization of multiple cameras and sensors are among the other practical challenges. The AI technique often referred as markerless pose estimation methods can overcome these challenges and aid biomechanists and clinicians. Thus, there is a need to develop new deep-learning models that can regress the musculoskeletal model directly from images and videos. However, the deep-learning models are dependent on the quality and quantity of training data. In the current scenario, training data for markerless pose estimation are dependent on the redundant marker-based systems and the challenges persist. To aid this, it is necessary to create a statistical human model or a skinned human animated motion from a biomechanical model to build more training data. From the skinned virtual data consisting of realistic movements, deep-learning models can be trained. Therefore, the aim of the research was to build a pipeline to develop a human-animated model from a musculoskeletal model i.e., the OpenSim model. Two different motions such as walking and running are illustrated as qualitative results. The gait pattern for walking and running motions are realistic from both the frontal and sagittal planes. Furthermore, the deep learning model (D3KE) built by Marian et. al was also evaluated on the animated human motions eg. walking motion from the above pipeline to validate the model. The performance of D3KE is evaluated from different planes of camera views and also a comparison between the upper and lower extremities. The evaluation and comparison are based on two metrics MAEangles (Mean Absolute Error of angles, in radians) and MPBLPE (Mean Per Bony Landmark Position Error, in cm). The MAEangles and MPBLPE are better when observed from the frontal plane rather than from the sagittal plane as the plane of view. Also, the joint angles in the upper extremity show better results compared to the lower extremity. Although, the predictions of the joint angles are way off from the ground truth. This opens the way to perform a feasibility study to optimize joint angles by a pixel loss refinement technique. The findings and remarks on the pixel-loss refinement is tabulated as results. ... Read more

At Basalt rehabilitation clinic The Hague, gait analysis is currently mostly observational. No objective measures are used and there is no systematic use of recordings. The aim of this research is to find and verify an accessible instrumented gait analysis method that could be implemented in the current care path at Basalt The Hague. The current instrumented gait analysis systems are marker-based. Recently more research has been done with regard to markerless systems, since markers bring along artefacts like the soft-tissue artefact. Markerless systems work with pose detection through artificial intelligence. A limitation of these pose detection systems, is that it is still in a research phase and did not make its way into the clinics yet. In a comparison of the accuracy of open-source pose estimation methods for measuring gait kinematics, OpenPose has been found most accurate.

For this research a program of requirements is made for implementation of instrumented gait analysis in the clinic at Basalt The Hague. The coordinates of anatomical landmarks obtained with OpenPose are processed to kinematic parameters. Besides, spatiotemporal parameters are looked into. A gait analysis set-up (consisting of tripods and mobile phone cameras) with OpenPose is proposed and tested to verify the accuracy. The OpenPose outcomes are compared with the the Simi motion capture system at Basalt Delft. By determining the Pearson correlation coefficients between OpenPose and Simi motion capture system, the kinematic parameters are verified. The mean standard deviations of the repeated recordings with OpenPose are used to determine the repeatability of OpenPose.

With this research is verified that OpenPose is repeatable and that kinematic parameters of the hip and knee are highly correlated to the standard method of instrumented gait analysis used at Basalt. Therefore can be concluded that OpenPose pose detection seems a promising method for determining kinematic and possibly also spatiotemporal parameters of gait. With follow-up research, especially clinical validation, and further development of the data processing, the first steps can be made towards implementation of accessible instrumented gait analysis in the current care path at Basalt The Hague, and possibly other locations and/or institutions. ... Read more

Humans perceive a pulling or pushing sensation when subjected to an asymmetric vibration. This so-called pseudo force has great potential to guide human movement. Previous research has exclusively focused on the effect of pseudo forces in open-loop environments, in which the user’s joint angular velocity cannot be corrected. As the latter is essential for providing movement guidance, this paper proposes the first closed-loop system in the field of pseudo forces, using amplitude-modulated pseudo forces as haptic feedback. With this feedback, the user was assisted in moving towards a specific target angular velocity. In a human factors experiment, the amplitude-modulated stimuli were compared to constant-amplitude stimuli. The results showed that amplitude-modulated pseudo forces significantly decreased the error between the user’s and the target angular velocity when continuous movement in the desired direction was achieved. Therefore, the study demonstrated that amplitude-modulated pseudo forces can effectively guide human movement, representing an essential step towards developing a wearable movement guidance device. ... Read more