Fall is a leading cause of death which suffers the elderly and society. Timed Up and Go (TUG) test is a common tool for fall risk assessment. In this paper, we propose a method for predicting TUG score from gait characteristics extracted from video based on computer vision and machine learning technologies. First, 3D pose is estimated from video captured with 2D and 3D cameras during human motion and then a group of gait characteristics are computed from 3D pose series. After that, copula entropy is used to select those characteristics which are mostly associated with TUG score. Finally, the selected characteristics are fed into the predictive models to predict TUG score. Experiments on real world data demonstrated the effectiveness of the proposed method. As a byproduct, the associations between TUG score and several gait characteristics are discovered, which laid the scientific foundation of the proposed method and make the predictive models such built interpretable to clinical users.

Falls are a major problem for older adults. A continuous unobtrusive in-home monitoring system that provides an accurate automated assessment of fall risk and detects when falls have occurred would allow for timely intervention and prevention allowing individual to remain healthier and independent longer. Sensor networks have been installed in apartments of older adult volunteers at TigerPlace, an independent senior living community. Initial results comparing gait parameters captured with a Microsoft Kinect with ground truth clinical fall risk assessments and GAITRite data are presented.