Prediction of Parkinson’s disease with various ML and DL techniques on speech data

Parkinson’s disease(PD) is a degenerative peripheral nervous system syndrome that influences muscle movement. It is incurable. Parkinson’s disease impacts more than one million people in India each year. Because of the disease’s degenerative nature, symptoms usually appear gradually and are barely perceptible, ranging from ordinary trembling of the hand to apparent speech, and voice, as well as writing alterations to serious symptoms such as loss of automatic movements. This research aims to contribute to the improvement of medical technology by assisting in the advanced and premature identification of PD, permitting experimental therapy by the use of a voice dataset from UCI Machine Learning(ML) database. We reviewed the current state of PD discernment in this study and discussed the prototype for quick identification of PD using several classifier propositions of ML and Deep Learning(DL). The best accuracy of about 94.8% was achieved by the usage of the CNN model with optimizer. We have also further built a web-based GUI for the web-based prediction of PD using model build.