Diagnosing Parkinson by using artificial neural networks and support vector machines

Abstract

Parkinson\'s Disease (PD) is the second most common neurodegenerative a�iction only surpassed by Alzheimer\'s Disease (AD). Moreover, it is expected to increase in the next decade with accelerating treatment costs as a consequence. This situation leads us towards the need to develop a Decision Support System for PD. In this paper we propose methods based on ANNs and SVMs to aid the specialist in the diagnosis of PD. Data recorded during 195 examinations carried out on 31 patients was used to verify the capacity of the proposed system. The results show a high accuracy of around 90%.