International Journal of Science, Engineering and TechnologyAuthors: Harini G, Sowmithra Murali
Abstract: Parkinson’s Disease (PD) is a progressive neurodegenerative disorder characterized by motor and non-motor symptoms that significantly impact a patient's quality of life. Conventional diagnostic methods, such as Magnetic Resonance Imaging (MRI) and Dopamine Transporter (DaT) scans, are expensive, time-consuming, and often inaccessible to individuals in low-resource settings. This research proposes an AI-powered multimodal detection system leveraging voice analysis, facial expression recognition, and movement tracking using smartphone sensors and low-cost wearables. The system integrates federated learning to enhance diagnostic accuracy while ensuring data privacy and scalability. Additionally, it incorporates real-time monitoring, environmental correlation, and blockchain-based health records to provide a comprehensive, cost-effective, and accessible solution for early PD detection. Our study estimates a cost reduction from ₹80,000–₹1,00,000 (MRI-based diagnosis) to ₹10,000–₹15,000 per device, making early screening feasible Our research focuses on developing an AI-powered smartphone-based detection system that can enable real-time screening and continuous monitoring, significantly improving early detection and patient outcomes.
