uTUG: An unsupervised Timed Up and Go test for Parkinson’s disease

Inertial measurement units (IMU) are used diagnostically in the movement analysis of Parkinson’s disease (PD) patients, allowing an objective way to assess biomechanical motion and gait parameters. The Timed Up Go (TUG) is a standardized clinical test widely monitoring patient fall risk progression. Gait tests performed at home have been applied as part protocols, enabling link supervised reference assessments. However, unsupervised real-world data context present challenges, mainly regarding interaction between participants recording system. Therefore, we developed evaluated novel algorithmic pipeline called TUG (uTUG). Our contribution automatic detection decomposition into their subphases, with no clinician supervision. In contrast related studies, only foot-mounted IMU additional markers or manual annotations, frames for subsequent classification by machine learning Support Vector Machine (SVM), Random Forest (RF) Naïve Bayes Classifier (NBC) algorithms. evaluation comprised 96 daily recordings 81 visits accumulating 300 real samples processed from 32 PD patients. A prefiltering sensitivity 98.6%, followed precision 90.6%, recall 88.5%, Fl-score 89.6% were achieved using RF continuous data. Thus, uTUG simplifies patients avoids annotations clinicians, automatically detecting tests.