Intelligent cyber-physical system for an efficient detection of Parkinson disease using fog computing

Parkinson’s disease is one of the notable neurodegenerative disorders caused by insufficient production of dopamine which damages the motor skills and voice. Advancement of the Internet of Things (IoT) has fuelled the development of healthcare systems. In this article, we propose an intelligent system for detecting Parkinson’s disease to provide proper medication by analysing voice samples. Instead of relying on limited storage capacity and computational resources of IoT, the recent healthcare systems take advantages of the cloud server. On the other hand, the utilization of cloud computing incurs the issues of data privacy and additional communication costs to the healthcare systems. To address this issue, we propose to utilize Fog computing as a midway layer between end devices and the cloud server. The proposed system employs the combinatorial Fuzzy K-nearest Neighbor and Case-based Reasoning classifier for the better classification of the Parkinson patients from healthy individuals. On the detection of abnormality, the proposed healthcare system is designed to generate an immediate alert to the patient. The proposed system is experimentally evaluated on the UCI-Parkinson dataset, and the results reveal the improved performance of our system over baseline approaches.