PARKINSONâS DISEASE CLASSIFICATION USING HYBRID NAMIB SQUIRREL SEARCH WATER ALGORITHM-BASED DEEP LEARNING APPROACH

Parkinson's disease (PD) is a neurological disease throughout the globe, and it is the second leading reason for death and impairment. The overall cases of PD have nearly doubled in the past 15 years. It has been defined by the absence of dopamine cells in the brain. PD affects about 1% of individuals over the age of 65, while 90% of them are affected by speech disorders like articulation, phonation, fluency, and prosody. Hence, the earlier prediction is significant in providing a good treatment for PD. In this research, the Namib squirrel search water algorithm (NSSWA) is proposed for PD classification. The voice sample is used as input and it is preprocessed using a Gaussian filter. Furthermore, feature extraction is applicable for the extraction of significant features, and the feature selection is done using the NSSWA. Moreover, the NSSWA-trained convolutional neural network (CNN) fused long short-term memory (LSTM) (CNN-LSTM), called NSSWA_CNN-LSTM, is used in PD classification. In addition, the efficacy of the model is validated via accuracy, sensitivity, specificity, loss function, mean-square error, and root-mean-square error with optimal values of 0.931, 0.934, 0.929, 0.068, 0.097, and 0.312 obtained.