Estimating spatiotemporal, kinematic, and kinetic movement variables with little obtrusion to the user is critical for clinical sports applications. One possible approach using a sparse inertial sensor setup, where sensors are not placed on all relevant body segments. Here, we investigated if can be estimated similarly accurate from setups as full lower-body setup. We by solving optimal control problems sagittal plane musculoskeletal models, in which minimized an objective that combined...

Machine learning is a promising approach to evaluate human movement based on wearable sensor data. A representative dataset for training data-driven models crucial ensure that the model generalizes well unseen However, acquisition of sufficient data time-consuming and often infeasible. We present method create realistic inertial with corresponding biomechanical variables by 2D walking running simulations. augmented measured simulated convolutional neural networks estimate sagittal plane...

This paper compares predictions of metabolic energy expenditure in gait using seven models to assess their correlation with experimental data. Ground reaction forces, marker data, and pulmonary gas exchange data were recorded for six walking trials at combinations two speeds, 0.8 m/s 1.3 m/s, three inclines, -8% (downhill), level, 8% (uphill). The cost, calculated the was compared cost from rates. A repeated measures showed that all correlated well correlations least 0.9. model by Bhargava...

Abstract Knee ligament sprains are common during change-of-direction (COD) maneuvers in multidirectional team sports. This study aimed to compare the effects of an 8-week injury prevention exercise program containing COD-specific exercises and a similar linear sprint on injury- performance-related variables 135° COD task. We hypothesized that training would lead (H1) stronger reductions biomechanical associated with anterior cruciate (ACL) risk COD, i.e. knee abduction moment angle, hip...

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 kinetics from data for a comprehensive biomechanical movement analysis is still challenging. Machine learning models or stepwise approaches performing Kalman filtering, inverse kinematics, dynamics can lead to inconsistencies between kinetics. We investigated reconstruction of arbitrary running...

Whether humans minimize metabolic energy in gait is unknown. Gradient-based optimization could be used to predict without using walking data but requires a twice differentiable model. Therefore, the model of Umberger et al. ( 2003 ) was adapted differentiable. Predictive simulations reaching task and were solved this continuous by minimizing effort. The simulation showed that minimization predicts unrealistic movements when compared effort minimization. predictive objectives other than are...

Trajectory optimization with musculoskeletal models can be used to reconstruct measured movements and predict changes in response environmental changes. It enables an exhaustive analysis of joint angles, moments, ground reaction forces, muscle among others. However, its application is still limited simplified problems two dimensional space or straight motions. The simulation directional changes, e.g. curved running, requires detailed three which lead a high-dimensional solution space. We...

Optimal control simulations of musculoskeletal models can be used to reconstruct motions measured with optical motion capture estimate joint and muscle kinematics kinetics. These are mutually dynamically consistent, in contrast traditional inverse methods. Commonly, optimal generated by tracking generalized coordinates combination ground reaction forces. The estimated from marker positions using, for example, kinematics. Hence, inaccuracies the tracked simulation. We developed an approach...

<title>Abstract</title> Currently, there is no established biomechanical model for surfing. Especially, musculoskeletal simulations can provide valuable insights athletes to enhance performance and prevent injuries using quantitative data such as joint angles, moments, muscle forces. The dynamic nature of surfing makes it difficult assess biomechanics in a laboratory setting. Although inertial measurement units (IMUs) offer potential solution, relying solely on IMU three-dimensional models...

This study aimed to develop and validate a machine learning method estimate continuous 3D knee moments during running from wearable sensor data. Reference were calculated 19 recreational runners treadmill at varying slopes (0 ± 5 % incline), speeds (self-selected 1 km/h) in 3 types of footwear. A convolutional neural network was trained on data 7 inertial measuring units (feet, shanks, thighs, sacrum) pair pressure insoles. We assessed performance over time windows (CONT) stance phases...

Postural instability represents one of the cardinal symptoms Parkinson's disease (PD). Still, internal processes leading to this are not fully understood. Simulations using neuromusculoskeletal human models could help understand these PD-associated postural deficits. In paper, we investigated whether reduced reactivity amplitudes resulting from impairments due PD can explain as well increased muscle tone often observed in individuals with PD. To simulate reactivity, gradually decreased...

Recent advances in wearable sensing and machine learning have created ample opportunities for "in the wild" movement analysis sports, since combination of both enables real-time feedback to be provided athletes coaches, as well long-term monitoring movements. The potential is useful performance enhancement or technique analysis, can achieved by training efficient models implementing them on dedicated hardware. Long-term used injury prevention, among others. Such applications are often...

Muscular co-contraction of antagonistic muscle pairs is often observed in human movement, but it considered inefficient and can currently not be predicted simulations where muscular effort or metabolic energy are minimized. Here, we investigated the relationship between minimizing systems with random uncertainty to see if minimize such system. We also effect time delay muscle, by varying neural control as well activation constant. solved optimal problems for a one-degree-of-freedom pendulum...

Humans control balance using different feedback loops involving the vestibular system, visual and proprioception. In this article, we focus on proprioception explore contribution of reflexes based force length to standing balance. particular, address questions how much alone could explain control, whether one modality, or feedback, is more important than other. A sagittal plane neuro-musculoskeletal model was developed with six degrees freedom nine muscles in each leg. controller designed...

Robotic exoskeletons have the potential to restore function in those with mobility impairments and enhance it able-bodied individuals. However, optimally controlling these devices, work concert human users, is challenging. Accurate model simulations of interaction between users may expedite design process improve control. Here, as a proof principle, we tested if could use predictive replicate gait adaptations changes energy expenditure from an experiment where participants walked...

Musculoskeletal simulations can be used to estimate biomechanical variables like muscle forces and joint torques from non-invasive experimental data using inverse forward methods. Inverse kinematics followed by dynamics (ID) uses body motion external force measurements compute movements the corresponding loads, respectively. ID leads residual (residuals) that are not physically realistic, because of measurement noise modeling assumptions. Forward dynamic (FD) found tracking data. They do...

Abstract Estimating spatiotemporal, kinematic, and kinetic movement variables with little obtrusion to the user is critical for clinical sports applications. Previously, we developed an approach estimate these from measurements seven lower-body inertial sensors, i.e., full setup, using optimal control simulations. Here, investigated if this similarly accurate when sparse sensor setups less sensors. To variables, solved problems on sagittal plane musculoskeletal models, in which objective was...