Parkinson's diagnosis using ant-lion optimisation algorithm

Abstract

Parkinson's disease (PD) is a long term progressive disorder of the central nervous system that mainly affects the movement of the body. But there are several limitations in detecting PD at an early stage. In this paper, a binary variant of the recently proposed ant-lion optimisation (ALO) algorithm has been proposed and implemented for diagnosing patients for Parkinson's disease at early stages. ALO is a recently proposed bio-inspired algorithm, which imitates the hunting patterns of ant-lions or doodlebugs proposed algorithm is used to find a minimum number of features that result in higher accuracy using machine learning classifiers. The proposed modified version of ALO extracts the optimal features for the two different Parkinson's Datasets with improved accuracy and computational time. The maximum accuracy achieved by the classifiers after optimal feature selection is 95.91%. The proposed algorithm results have been compared with other related algorithms for the same datasets.