Research Projects

Musculoskeletal and neuromuscular health conditions, such as osteoarthritis, cerebral palsy, and sports injuries, affect an individual’s mobility, strength, and coordination. In such cases, rehabilitation is crucial for restoring functionality, improving physical abilities, and enhancing quality of life. However, progress in rehabilitation is often monitored using subjective methods, such as observational movement evaluation, patient-reported outcomes, and daily activity logs. While these methods provide qualitative insights into patient improvement, they are inherently limited by observer’s bias and variability in expertise. To address these limitations, there is a need for objective, quantifiable methods that can reliably track functional improvements and enable clinicians to tailor interventions more effectively. Kinesiology, the scientific study of human movement, can address these needs by providing quantifiable methods to monitor movement. Skin marker-based optoelectronic motion capture (MoCap) systems are considered the gold standard for obtaining objective data, such as joint kinematics and spatiotemporal parameters (e.g., cadence, stride length, step width). Despite their accuracy, these systems are expensive, require skilled operation, and are restricted to controlled lab settings. Markerless video-based systems provide a promising alternative, using 2D video data and deep learning algorithms to estimate 3D motion unobtrusively and cost-effectively. Their accessibility and ease of use make them ideal for clinical and rehabilitation applications, although their current application is still in its infancy. The global aim of this project is to democratize motion analysis by developing markerless video-based systems for clinical applications targeting musculoskeletal and neuromuscular conditions. The project includes two experimental phases. First, controlled studies will compare the accuracy of marker-based and markerless systems in evaluating joint angles and spatiotemporal parameters during activities such as walking, sit-to-stand transitions, stair navigation, and dynamic tasks. Second, motion data from both systems will be integrated with musculoskeletal models to simulate and compare joint kinetics, such as forces and moments. The study will recruit both healthy participants and patients with musculoskeletal and neuromuscular conditions. The proximity of the Institute of Medical Sciences (BHU) to IIT-BHU offers a unique advantage for patient recruitment and trials. This project aims to bridge the gap between advanced, yet complex and expensive, motion analysis technologies and practical clinical applications. By advancing markerless systems, it seeks to make motion analysis accessible in hospital and rehabilitation settings, transforming patient monitoring and treatment planning. The expected outcomes will significantly enhance the applicability of motion analysis tools in rehabilitation biomechanics and sports science.