Automatic Timed Up-and-Go Sub-Task Segmentation for Parkinson’s Disease Patients Using Video-Based Activity Classification

摘要: The timed up-and-go (TUG) test has been widely accepted as a standard assessment for measuring the basic functional mobility of patients with Parkinson’s disease. Several sub-tasks “Sit,” “Sit-to-Stand,” “Walk,” “Turn,” “Walk-Back,” and “Sit-Back” are included in TUG test. It shown that time costs these useful clinical parameters automatic methods have proposed to segment However, usually require either well-controlled environments video recording or information from special devices, such wearable inertial sensors, ambient depth cameras. In this paper, an sub-task segmentation method using video-based activity classification is validated study 24 disease patients. Videos used paper recorded semi-controlled various backgrounds. state-of-the-art deep learning-base 2-D human pose estimation technologies feature extraction. A support vector machine long short-term memory network then subtask segmentation. Our can be automatically acquire videos-only, leading possibility remote monitoring patients’ condition.