Patient Questionnaires Based Parkinson’s Disease Classification Using Artificial Neural Network

Parkinson’s disease is one of the most prevalent and harmful neurodegenerative conditions (PD). Even today, PD diagnosis monitoring remain pricy inconvenient processes. With unprecedented progress artificial intelligence algorithms, there an opportunity to develop a cost-effective system for diagnosing at earlier stage. No permanent remedy has been established yet; however, helps lead better life. Probably, three responsible categories symptoms Disease are tremors, rigidity, body bradykinesia. Therefore, we investigate 53 unique features Progression Markers Initiative dataset determine significant symptoms, including major categories. As feature selection integral developing generalized model, excluding selection. Four methods incorporated—low variance filter, Wilcoxon rank-sum test, principle component analysis, Chi-square test. Furthermore, utilize machine learning, ensemble neural networks (ANN) classification. Experimental evidence shows that not all equally important, but no symptom can be completely eliminated. However, our proposed ANN model attains best mean accuracy 99.51%, 98.17% specificity, 0.9830 Kappa Score, 0.99 AUC, 99.70% F1-score with features. The efficiency suggested technique on diverse data modalities demonstrated by comparison recent publications. Finally, trade-off between classification time accuracy.