A Comparative and Explainable Study of Machine Learning Models for Early Detection of Parkinson's Disease Using Spectrograms

Hadjer Zebidi, Zeineb BenMessaoud, Mondher Frikha

2025

Abstract

Parkinson's disease (PD) is a progressive neurodegenerative disorder that originally affects the motor system. Therefore, early diagnosis is essential for effective intervention. Classic diagnostic approaches heavily rely on clinical observations and manual feature extraction, limiting the detection of subtle early vocal impairments. This research examines machine learning (ML) techniques, namely Support Vector Machines (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), for early identification of PD through the analysis of spectrogram images derived from voice recordings. Mel-Frequency Cepstral Coefficients (MFCC), Short-Time Fourier Transform (STFT), and Mel-Spectrograms were extracted. The improvement of the model was introduced by the Synthetic Minority Over-sampling Technique (SMOTE) and hyperparameter tuning using GridSearchCV (Grid Search with Cross-Validation). Implementing the above methods resulted in significant performance improvements, with XGBoost achieving an accuracy of 95 ± 0.02 on the PC-GITA dataset and SVM attaining 90.74 ± 0.04 on the Neurovoz dataset. Local Interpretable Model-agnostic Explanations (LIME) enhanced model transparency by identifying the significant regions in spectrograms that most influence predictions. This analysis illustrates the efficacy of ML models utilizing SMOTE and GridSearchCV, particularly when augmented by LIME for interpretability, in improving early detection of PD, thereby presenting a feasible approach for clinical implementation.

Paper Citation

in Harvard Style

Zebidi H., BenMessaoud Z. and Frikha M. (2025). A Comparative and Explainable Study of Machine Learning Models for Early Detection of Parkinson's Disease Using Spectrograms. In Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-730-6, SciTePress, pages 272-282. DOI: 10.5220/0013183900003905

in Bibtex Style

@conference{icpram25,
author={Hadjer Zebidi and Zeineb BenMessaoud and Mondher Frikha},
title={A Comparative and Explainable Study of Machine Learning Models for Early Detection of Parkinson's Disease Using Spectrograms},
booktitle={Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2025},
pages={272-282},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013183900003905},
isbn={978-989-758-730-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - A Comparative and Explainable Study of Machine Learning Models for Early Detection of Parkinson's Disease Using Spectrograms
SN - 978-989-758-730-6
AU - Zebidi H.
AU - BenMessaoud Z.
AU - Frikha M.
PY - 2025
SP - 272
EP - 282
DO - 10.5220/0013183900003905
PB - SciTePress