Parkinson’s Disease Prediction Using Machine Learning and Nature-Inspired Optimization Technique

Parkinson’s disease (PD) is a nervous system condition that gradually becomes worse with the passing of time and is caused by injury to brain cells responsible for producing dopamine. When between 60 and 80% of dopamine-producing are lost, the production of dopamine diminishes, leading to the manifestation of motor symptoms associated with PD. Artificial intelligence (AI), particularly machine learning (ML), has become increasingly pivotal in medicine. Optimization techniques play a vital role in enhancing the accuracy and efficiency of disease prediction models. In the proposed work, we have implemented a nature-inspired firefly optimization technique to see its results for the correct disease classification. It is implemented using KNN, SVM, decision tree, and random forest combined with a firefly algorithm for the classification while minimizing the features selected. The highest results achieved are using random forest as a classifier, with a maximum accuracy of 87.22%.