INERTIA SENSOR-BASED MOBILITY ANALYSIS FOR PARKINSON'S DISEASE

Abstract

Movement disorders such as Parkinson’s disease (PD) will affect a rapidly growing segment of the population as society continues to age. Without a known cure, PD is causing tremendous financial and logistical challenges for patients and society. The early diagnosis of PD and accurate assessment of PD severity stage is crucial for prompt and proper treatment. However, it currently requires a skilled doctor and specialized equipment, and it is expensive, non-scalable and prone to human error. In this thesis, we developed a cost-effective, scalable, objective and accurate system for mobility analysis and PD diagnosis. The system uses custom-built wearable inertial sensors to monitor a person’s movement such as walking and hand tremor. Novel algorithms were developed to quantify movement parameters (e.g., step time and step length) and classify the severity stage of PD. The practicality and accuracy of the system has been validated in clinical studies involving more than 80 PD patients and control subjects.