期刊
WIRELESS MOBILE COMMUNICATION AND HEALTHCARE
卷 247, 期 -, 页码 12-19出版社
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-319-98551-0_2
关键词
Sensors; Activity recognition; Smartphone; Accelerometer; Machine learning; Deep learning; Parkinson's disease; Clinical trial
Smartphone-based assessments have been considered a potential solution for continuously monitoring gait and mobility in mild to moderate Parkinson's disease (PD) patients. Forty-four PD patients from cohorts 4 to 6 of the Multiple Ascending Dose (MAD) study of PRX002/RG7935 and thirty-five age-and gender-matched healthy individuals (i.e. healthy controls - HC) in a separate study performed smartphone-based assessments for up to 24weeks and up to 6 weeks, respectively. The assessments included active gait tests, where all participants were asked to walk for 30 s with at least one 180. turn, and passive monitoring, in which subjects carried the smartphone in a pocket or fanny pack as part of their daily routine. In total, over 6,600 active gait tests and over 30,000 h of passive monitoring data were collected. A mobility analysis indicates that patients with PD are less mobile than HCs, as manifested in time spent in gait-related activities, number of turns and sit-to-stand transitions, and power per step. It supports the potential use of smartphones for continuous mobility monitoring in future clinical practice and drug development.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据