Tian Tan

This paper develops deep learning models to detect neuromuscular diseases and quantify movement impairment using video-derived biomechanics data from smartphone recordings.

This paper develops a hybrid machine learning and musculoskeletal simulation framework to improve dynamics estimation from smartphone video during walking, achieving lower error rates than simulation-only or ML-only approaches.

This paper uses self-supervised learning to improve accuracy and reduce data requirements for estimating ground reaction forces from IMU data.