Smartphone-Based Self-Diagnosis of Parkinson’s Disease

Abstract Parkinson's Disease (PD) is a degenerative nervous system disorder that affects the motor system progressively over a long period of time leading to shaking, body stiffness, and difficulty with walking, balance, and coordination. Over 60,000 people are diagnosed with PD every year in America alone. This number is expected to amount to 1.2million people living with PD by the year 2030. PD is an incurable disease, but it can be managed efficiently by medication, therapy, and or surgery if diagnosed early to help relieve symptoms and maintain a good quality of life. However, there is no reliable, easily accessible, and affordable testing procedure for the patients to self-administer for detecting early onset of PD. The main goal of this project was to develop a smartphone-based easy-to-use self-diagnostic tool to detect early stages of PD while creating awareness about its symptoms and the importance of seeking early intervention. This work uses a cross-platform mobile application development environment to develop an android and iOS compatible smartphone app. The app uses smartphone's built-in hardware sensors such as accelerometer and gyroscope to measure and analyse the frequency of body tremors (4-6 Hz range), using Fast Fourier Transformations (FFT). The app also utilizes a specially designed spiral test to analyse the user's tracing of an Archimedean spiral which is another indicator of deteriorating motor functions in PD. These two tests combined with the user information collected from a demographic questionnaire probing various risk factors of PD are used to estimate a likely occurrence of PD, suggesting the user to seek medical assistance. Our project exploits the ubiquitous smartphone technology to provide user-friendly and affordable means to self-monitor the symptoms of Parkinson's disease and create a much-needed societal awareness about the disease.