Abstract:
This research describes the design and development of a machine learning system for
detecting Parkinson's disease. The system includes Python method creation, support vector
machine (SVM) classifier building, and data analysis utilising UCI/Oxford university
datasets[0]. Pitch, jitter, shimmer, and harmonic-to-noise ratio are among the variables
retrieved from speech recordings of healthy and Parkinson's patients in the dataset. The
technology also includes a mobile application that can detect Parkinson's disease from
speech recordings using the classifier. Users can record their voice samples and save them
locally on the device using the mobile application. The system is built on a Raspberry Pi
device, which includes a microphone module that can gather voice signals and execute the
classifier. The project's goal is to develop a low-cost, portable, and accurate Parkinson's
disease diagnosis tool that may be used by anybody, anywhere
