Clinical decision support system for Parkinson's disease and related movement disorders

Abstract

The goal of this study is to accurately distinguish between finger tremors of Parkinson's disease and other movement disorders using a tri-axial gyroscope. Finger tremor is specifically studied here as compared to hand tremor since farther distance from radio-carpal joint results in better acquisition of tremor signal. This study is an effort towards providing physicians with a clinical decision support system to facilitate them in accurate diagnosis of Parkinson's disease (PD) and help them in discriminating between other movement disorders and PD. We designed the hardware to acquire angular displacement from tri-axial gyroscope and apply a series of techniques to extract different features in time and frequency domains. Both resting and postural tremor is studied for analysis. A total of 104 people participated in our study, using features from this data we were able to create a Clinical Decision Support System (CDSS) with overall accuracy of 82.43%. Additionally, CDSS was also used as a first hand diagnostic tool in a real hospital setting with an accuracy of 77.78%.