Author:

Keywords:

Artificial intelligence, Biomechanics, Computational models, Musculoskeletal, Science & Technology, Life Sciences & Biomedicine, Technology, Biophysics, Engineering, Biomedical, Engineering, TIBIOFEMORAL CONTACT FORCES, VIRTUAL PHYSIOLOGICAL HUMAN, ESTIMATE MUSCLE FORCES, FINITE-ELEMENT-ANALYSIS, JOINT MOMENTS, MUSCULOSKELETAL MODEL, KNEE-JOINT, PRECISION-MEDICINE, BIG DATA, IN-VIVO, Biomechanical Phenomena, Humans, Imaging, Three-Dimensional, Machine Learning, Models, Anatomic, Musculoskeletal System, Nervous System, 0903 Biomedical Engineering, 0913 Mechanical Engineering, Biomedical Engineering, 4003 Biomedical engineering

Abstract:

Many biomedical, orthopaedic, and industrial applications are emerging that will benefit from personalized neuromusculoskeletal models. Applications include refined diagnostics, prediction of treatment trajectories for neuromusculoskeletal diseases, in silico design, development, and testing of medical implants, and human-machine interfaces to support assistive technologies. This review proposes how physics-based simulation, combined with machine learning approaches from big data, can be used to develop high-fidelity personalized representations of the human neuromusculoskeletal system. The core neuromusculoskeletal model features requiring personalization are identified, and big data/machine learning approaches for implementation are presented together with recommendations for further research.