Prediction of Parkinson's disease using speech signal with Extreme Learning Machine

Speech impairments analysis has been used as an efficient tool for early detection of Parkinson's disease (PD). In this paper, we have proposed an efficient approach using Extreme Learning Machine to predict Parkinson's disease accurately utilising speech samples. The performance of the method has been assessed with a reliable dataset from UCI repository. The proposed method distinguishes Parkinson diseased subjects and healthy subjects with an accuracy of 90.76% and 0.81 MCC for the training dataset. When tested with an independent dataset comprising of Parkinson diseased patients, the proposed method gives 81.55% accuracy. The performance of our method is compared with existing techniques such as Neural Network and Support Vector Machine. The results obtained depict that the proffered method is reliable for identifying the Parkinson's disease.