ARTORG Center for Biomedical Engineering Research

Motor Symptom Classification of Parkinson’s Disease using a Comprehensive Home-Based Sensor System

Project Members:  Angela Botros, Prabitha Urwyler, Martin Camenzind, Daniel Bolliger, Tim Vanbellingen, Rolf Kistler, Stephan Bohlhalter, Tobias Nef
Project Start: 01.09.2017
Project Funding: Hasler Foundation

Parkinson‘s disease (PD) is one of the most common neurodegenerative diseases affecting more than 4.1 million patients worldwide and about 15’000 patients in Switzerland alone. Since PD is age-related, the number of patients is expected to double by 2030 following demographic trends. The cardinal PD symptoms are movement-related, including resting tremor, rigidity, bradykinesia, postural instability and freezing of gait.

There is a strong clinical need for objective and continuous monitoring of motor performance in PD for improving therapeutic regiments and for assessment in clinical trials. In recent years, there has been a lot of progress in the area of symptom identification in PD patients. Most of the developed systems are based on a set of wearable sensors which classify the symptoms using a variety of machine learning algorithms and signal processing techniques. The majority of the approaches focus on one symptom only, e.g. Freezing of gait or Bradykinesia, in a clinical setting.

And while these approaches are able to identify the symptoms with high sensitivity and specificity, usually the underlying hardware is not designed for home- and long-term usage. The aim of this project is to develop a new system for the monitoring of PD motor symptoms, intended for long-term usage in a non-clinical environment, i.e. at the peoples homes. Key values are the simplicity, patient acceptance and robustness of the system.

Keywords: Parkinson's Disease, motor symptoms, wearable sensors, monitoring