Diagnostic and Support Information System for Parkinson's Disease Patients

Abstract

This study outlines a deployment strategy for a Parkinson's disease diagnostic model within the Internet of Things (IoT) ecosystem. It focuses on the collaborative operation of system components for data processing and storage. The IoT device captures data via sensors and smartphones, initiating data processing and feature extraction. Data is then routed through the Local Flask Server to the Open Semantic Technology for Intelligent Systems (OSTIS) Server, a knowledge graph platform that processes and interprets the data. A Neural Network Predictor Agent within the OSTIS Server manages neural network models, executing them to generate predictions linked to the system's knowledge base, ultimately stored in the local database. This process involves data acquisition and preprocessing by the IoT device, transmission to the Flask server, and further processing by the OSTIS server, leading to decision-making support. This architecture enables real-time analysis and complex pattern recognition, integrating data with a knowledge graph for advanced analysis and decisions, ensuring efficient operation and management of each component. The long-term management of Parkinson’s disease requires continuous, adaptive, and individualized therapy. The IT-therapy module introduces powered by a neural network-based digital twin, which extends the diagnostic system into a therapeutic decision-support framework.