Performance Comparison of Cuckoo Search and Ant Colony Optimization for Identification of Parkinson’s Disease Using Optimal Feature Selection

Parkinson’s disease (PD) is a chronic central nervous system condition that largely affects the body movement. If left untreated at a nascent stage, it could be life threatening. PD leads to sluggishness of movement and causes muscle inflexibility and tremors. There are numerous methods available in literatures such as speech-based method, gait-based methods, handwriting-based methods for detection of PD. However, speech-based method is known to be an efficient and competitive method as compared other two methods. Hence, in this study, we consider speech-based method in which two nature-inspired algorithm, viz., cuckoo search and ant colony optimization (ACO) have been used to select the optimal features for classification of the Parkinson patient with respect to healthy ones. The simulation results reveal that the cuckoo search algorithm obtained the better accuracy and achieved minimal subset of features with more stability compared to ACO algorithm.