Automatic and Early Detection of Parkinson's Disease by Analyzing Acoustic Signals Using Classification Algorithms Based on Recursive Feature Elimination Method

Aug 2, 2026

Parkinson's disease (PD) is a neurodegenerative condition driven by impaired dopamine production, often leading to delayed clinical diagnosis through extensive physical, psychological, and specialist examinations. This study develops an early-detection methodology that analyzes voice recordings by extracting informative features and applying machine-learning models to distinguish Parkinson’s cases from healthy subjects. The approach optimizes feature selection via recursive feature elimination (RFE) with hyperparameter tuning and uses SMOTE to balance the dataset, achieving strong classification performance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages24

Reading Time1 h 0 min

Abstract

Introduction

Background of Parkinson's disease

Symptoms and clinical challenges

Motivation for voice-based detection

Materials and Methods

Results and Discussion

Conclusion

diagnostics Article Automatic and Early Detection of Parkinson's Disease by Analyzing Acoustic Signals Using Classiﬁcation Algorithms Based on Recursive Feature Elimination Method Khaled M. Alalayah 1, *, Ebrahim Mohammed Senan 2, *, Hany F. Atlam 3, Ibrahim Abdulrab Ahmed 4 and Hamzeh Salameh Ahmad Shatnawi 4 Citation: Alalayah, K.M.; Senan, E.M.; Atlam, H.F.; Ahmed, I.A.; Shatnawi, H.S.A. Automatic and Early Detection of Parkinson's Disease by Analyzing Acoustic Signals Using Classiﬁcation Algorithms Based on Recursive Feature Elimination Method. Diagnostics 2023, 13, 1924 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)diagnostics13111924 Academic Editor: Mehedi Masud Received: 11 May 2023 Revised: 23 May 2023 Accepted: 27 May 2023 Published: 31 May 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Department of Computer Science, Faculty of Science and Arts, Najran University, Sharurah 68341, Saudi Arabia 2 Department of Artiﬁcial Intelligence, Faculty of Computer Science and Information Technology, Alrazi University, Sana'a, Yemen 3 Cyber Security Centre, WMG, University of Warwick, Coventry CV4 7AL, UK; [hany.atlam@warwick.ac.uk](hany.atlam@warwick.ac.uk) 4 Computer Department, Applied College, Najran University, Najran 66462, Saudi Arabia; [iaalqubati@nu.edu.sa](iaalqubati@nu.edu.sa) (I.A.A.); [hsshatnawi@nu.edu.sa](hsshatnawi@nu.edu.sa) (H.S.A.S.) * Correspondence: [kmalalayah@nu.edu.sa](kmalalayah@nu.edu.sa) (K.M.A.); [senan26102020@gmail.com](senan26102020@gmail.com) (E.M.S.) Abstract: Parkinson's disease (PD) is a neurodegenerative condition generated by the dysfunction of brain cells and their 60–80% inability to produce dopamine, an organic chemical responsible for controlling a person's movement. This condition causes PD symptoms to appear. Diagnosis involves many physical and psychological tests and specialist examinations of the patient's nervous system, which causes several issues. The methodology method of early diagnosis of PD is based on analysing voice disorders. This method extracts a set of features from a recording of the person's voice. Then machine-learning (ML) methods are used to analyse and diagnose the recorded voice to distinguish Parkinson's cases from healthy ones. This paper proposes novel techniques to optimize the techniques for early diagnosis of PD by evaluating selected features and hyperparameter tuning of ML algorithms for diagnosing PD based on voice disorders. The dataset was balanced by the synthetic minority oversampling technique (SMOTE) and features were arranged according to their contribution to the target characteristic by the recursive feature elimination (RFE) algorithm. We applied two algorithms, t-distributed stochastic neighbour embedding (t-SNE) and principal component analysis (PCA), to reduce the dimensions of the dataset. Both t-SNE and PCA ﬁnally fed the resulting features into the classiﬁers support-vector machine (SVM), K-nearest neighbours (KNN), decision tree (DT), random forest (RF), and multilayer perception (MLP) . Experimental results proved that the proposed techniques were superior to existing studies in which RF with the t-SNE algorithm yielded an accuracy of 97%, precision of 96.50%, recall of 94%, and F1-score of 95% . In addition, MLP with the PCA algorithm yielded an accuracy of 98%, precision of 97.66%, recall of 96%, and F1-score of 96 .66% . Keywords: Parkinson's disease; exploratory data analysis; coefﬁcient of variation; t-SNE; REF; machine learning 1. Introduction Parkinson's disease (PD) is a neurodegenerative disease caused by the death of neurons (called substantia

FAQs

Why is early detection of Parkinson's disease important?

Early diagnosis can reduce reliance on complex specialist evaluations by using accessible voice disorder signals that correlate with PD-related vocal impairment.

How does the proposed method detect Parkinson's disease?

The method extracts features from voice recordings, applies RFE to select features, balances data with SMOTE, reduces dimensionality using t-SNE or PCA, and then classifies using models such as SVM, KNN, decision tree, random forest, and MLP.

Which models and dimensionality-reduction methods achieved the best reported performance?

Random forest combined with t-SNE achieved high accuracy (97%), and MLP combined with PCA achieved even higher accuracy (98%), according to the reported metrics.

Aug 2, 2026

Upload by Olivia Brown