Parkinson’s disease Prediction using Quasi Optimal Optimization Algorithm over Big Data

Abstract

Parkinson’s Disease (PD) is a neurodegenerative disorder of the central nervous system of people worldwide. It can affect mostly the motor functions. The PD is observed by bradykinesia, rigidity, resting tremor, postural instability, sleeping problems, speech problem, and disordering of the vocal cord at an early stage. The voice disorders the PD patients more than 90%.  If the disease is predicted at an early stage, then the doctor can decide to give treatment for increasing the patient’s living period. Here, we aim that to predict PD using patient voice recording data set using Big Data Analytics (BDA). In our approach, we propose a disease prediction model that uses machine learning-based classifier algorithms such as Logistic Regression (LR), Naïve Bayes (NB), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Neural Network (NN), and Algorithm Quasi (AQ). The result shows an average accuracy of 96.66. The recorded voices of patients are converted to voice parameters like jitter, shimmer, Harmonic to Noise Ratio (HNR), Recurrence Period Density Entropy (RPDE), Detrended Fluctuation Analysis (DFA), Pitch Period Entropy (PPE), and Unified Parkinson's Disease Rating Scale (UPRS) by using R Programming. The status of Parkinson’s Disease is found based on testing patient voice data set whether a person has Parkinson’s disease or not.