A mobile game for Parkinson's disease symptom detection

• Main Author: Chan, Yong Lin
• Other Authors: Yu Han
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/171985
• Institution: Nanyang Technological University

Parkinson’s Disease (PD) is a chronic disease where symptoms worsen over time. Early detection of the disease allows effective treatments and help to improve the patients’ daily lives. With a digital PD screening tool, it is useful for people to assess their condition easily at their convenience. This paper introduces a prototype which aims to detect signs of PD from several segments on a mobile application. The segments include audio recording, spiral drawing, trail making tests, and picture recognition. The prototype collects audio features and spiral drawings, and models were built to assess users’ PD risk. The audio model achieved a 92.3% accuracy, and the spiral image model attained an 83% accuracy. In the future, collaborating with medical centres for data collection can improve the models’ accuracy.