Speech-Based Ensemble Method For Detecting Parkinson’s Disease
DOI:
https://doi.org/10.29284/5bf42x07Keywords:
Ensemble Learning, Machine Learning, Parkinson's Disease, Speech Impairment, Speech Signals.Abstract
Parkinson's disease (PD) is ranked next to the Alzheimer as the most common neurodegenerative disease. PD is a progressive disease which is incurable as yet. Early detection of PD is vital to slow down its progress. About 90% of the PD patients suffer from speech impairment. Speech is a crucial biomarker for the early detection of PD. Speech signals can be effectively utilized in machine learning to predict the disease in non-invasive and cost-effective way. In several studies of machine learning, effort has been made to predict PD. But, in most of the studies issues like data preprocessing, data imbalance, and feature selection are often overlooked. In this study, different ensemble techniques like Random Forest, Bagging with various base estimators, GradientBoosting, XGBoost, AdaBoost, Stacking and Voting are applied to speech features. Class imbalance in a dataset causes machine learning algorithms to be biased to the majority class, leading to the misclassification problem. This issue is addressed by applying data balancing techniques. The different approaches to data balancing have adopted different algorithms for producing a balanced dataset. As a result, different data balancing techniques have different impacts on the machine learning algorithms with the same dataset, producing varying classification accuracy. Hence, different data balancing techniques are explored in this study to obtain 98.31% of accuracy. ANOVA F-test is performed for feature selection. 118 machine learning models are tested. Model performance is measured in terms of accuracy, sensitivity, specificity, F1-score, execution Time and ROC curve.
References
[1] R. A. Shirvan and E. Tahami, “Voice Analysis for Detecting Parkinson’s Disease Using Genetic Algorithm and KNN Classification Method,” no. December, pp. 14–16, 2011.
[2] Ahlrichs, C., and Lawo, M. (2013). Parkinson’s disease motor symptoms in machine learning: a review. arXiv preprint arXiv:1312.3825. doi: 10.5121/hiij.2013.2401
[3] Tsanas, A., Little, M. A., McSharry, P. E., Spielman, J., and Ramig, L.O. (2012). Novel speech signal processing algorithms for high-accuracy classification of Parkinson’s disease. IEEE Trans. Bio. Med. Eng. 59, 1264–1271. doi: 10.1109/TBME.2012.2183367
[4] Harel B, Cannizzaro M, Snyder PJ. Variability in fundamental frequency during speech in prodromal and incipient Parkinson’s disease: A longitudinal case study. Brain and cognition. 2004 Oct 31; 56 (1):24–9.
https://doi.org/10.1016/j.bandc.2004.05.002 PMID: 15380872
[5] Tsanas A, Little MA, McSharry PE, Ramig LO. Accurate telemonitoring of Parkinson’s progression by noninvasive speech tests. IEEE transactions on Biomedical Engineering. 2010 Apr; 57(4):884– 93. https://doi.org/10.1109/TBME.2009.2036000 PMID: 19932995
[6] Palma, J.A., Kaufmann, H.: Autonomic disorders predicting Parkinson disease. Parkinsonism Relat Disord. 2014 January; 20(0 1): S94–S98. doi:10.1016/S1353 8020(13)70024-5.
[7] Chaudhuri, K.R., Healy, D.G., Schapira, A.H.V.: Non-motor symptoms of Parkinson’s disease: diagnosis and management. Lancet Neurol 2006; 5: 235–45
[8] World Health Org. Parkinson Disease. Accessed: Nov. 20, 2025. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/parkinson Disease
[9] Castillo-Barnes, D., Ramírez, J., Segovia, F., Martínez-Murcia, F. J., Salas-Gonzalez, D., and Górriz, J. M. (2018). Robust ensemble classification methodology for I123-Ioflupane SPECT images and multiple heterogeneous biomarkers in the diagnosis of Parkinson’s disease. Front. Neuroinf. 12:53. doi: 10.3389/fninf.2018.00053
[10] Adeli, E., Shi, F., An, L., Wee, C.-Y., Wu, G., Wang, T., et al. (2016). Joint feature-sample selection and robust diagnosis of Parkinson’s disease from MRI data. NeuroImage 141, 206–219. doi: 10.1016/j.neuroimage.2016.05.054
[11] Xu, N., Zhou, Y., Patel, A., Zhang, N., Liu, Y. Parkinson’s Disease Diagnosis beyond Clinical Features: A Bio-marker using Topological Machine Learning of rs-fMRI. Available: https://ssrn.com/abstract=4165341
[12] Abujrida, H., Agu, E., and Pahlavan, K. (2017). “Smartphone-based gait assessment to infer Parkinson’s disease severity using crowdsourced data,” in 2017 IEEE Healthcare Innovations and Point of Care Technologies (HI-POCT) (Bethesda, MD), 208–211. doi: 10.1109/HIC.2017.8227621
[13] Adams, W. R. (2017). High-accuracy detection of early Parkinson’s Disease using multiple characteristics of finger movement while typing. PLoS ONE 12: e0188226. doi: 10.1371/journal.pone.0188226
[14] Al-Fatlawi, A. H., Jabardi, M. H., and Ling, S. H. (2016). “Efficient diagnosis system for Parkinson’s disease using deep belief network,” in 2016 IEEE Congress on Evolutionary Computation (CEC) (Vancouver, BC), 1324–1330. doi: 10.1109/CEC.2016.7743941
[15] UCI Machine Learning Repository, website: https://archive.ics.uci.edu/dataset/174/parkinsons
[16] 'Exploiting Nonlinear Recurrence and Fractal Scaling Properties for Voice Disorder Detection', Little MA, McSharry PE, Roberts SJ, Costello DAE, Moroz IM.BioMedical Engineering OnLine 2007, 6:23 (26 June 2007)
[17] K. M. Alalayah, E. M. Senan, H. F. Atlam, I. A. Ahmed, and H. S. A. Shatnawi, ‘‘Automatic and early detection of Parkinson’s disease by analyzing acoustic signals using classification algorithms based on recursive feature elimination method,’’ Diagnostics, vol. 13, no.11, p. 1924, May 2023.
[18] R. Aishwarya, K. Pavitra, P. V. Miranda, K. Keerthana, and L. Kamatchi Priya, ‘‘Parkinson’s disease prediction using Fisher score based recursive feature elimination,’’ in Proc. Int. Conf. Advancement Comput. Comput. Technol. (InCACCT), May 2023, pp. 1–8.
[19] Shyamala, K., and Navamani, T.M. Design of an Efficient Prediction Model for Early Parkinson’s Disease Diagnosis. IEEE Access, vol. 12, Oct. 2024. doi: 10.1109/ACCESS.2024.3421302
[20] Rahman,S.; Hasan, M.; Sarkar, A.K.; Khan, F. Classification of Parkinson’s disease using speech signal with machine learning and deep learning approaches. Eur. J. Electr. Eng. Comput. Sci. 2023, 7, 20–27.
[21] Mondol,S.R., Kim, R.; Lee, S. Hybrid machine learning framework for multistage Parkinson’s disease classification using acoustic features of sustained korean vowels. Bioengineering 2023, 10, 984.
[22] Lamba,R.; Gulati, T., Alharbi, H.; Jain, A. A hybrid system for Parkinson’s disease diagnosis using machine learning techniques. Int. J. Speech Technol. 2022, 25, 583–593.
[23] Bukhari, S.N.H.; Ogudo, K.A. Ensemble Machine Learning Approach for Parkinson’s Disease Detection Using Speech Signals. Mathematics 2024, 12, 1575. https:// doi.org/10.3390/math12101575
[24] Di Cesare, M.G.; Perpetuini, D.; Cardone, D.; Merla, A. Machine Learning-Assisted Speech Analysis for Early Detection of Parkinson’s Disease: A Study on Speaker Diarization and Classification Techniques. Sensors 2024, 24, 1499. https://doi.org/10.3390/s24051499
[25] Ma, H., Tan, T., Zhou, H., and Gao, T. (2016). “Support Vector Machine- recursive feature elimination for the diagnosis of Parkinson disease based on speech analysis,” in 2016 Seventh International Conference on Intelligent Control and Information Processing (ICICIP) (Siem Reap), 34–40. doi: 10.1109/ICICIP.2016.7885912
[26] Bakar, Z. A., Ispawi, D. I., Ibrahim, N. F., and Tahir, N. M. (2012). “Classification of Parkinson’s disease based on Multilayer Perceptrons (MLPs) neural network and ANOVA as a feature extraction,” in 2012 IEEE 8th International Colloquium on Signal Processing and its Applications) (Malacca), 63–67. doi: 10.1109/CSPA.2012.6194692
[27] Haq, A. U., Li, J., Memon, M. H., Khan, J., Din, S. U., Ahad, I., et al. (2018). “Comparative analysis of the classification performance of machine learning classifiers and deep neural network classifier for prediction of Parkinson disease,” in 2018 15th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) (Chengdu), 101–106.
[28] Islam, M. S., Parvez, I., Hai, D., and Goswami, P. (2014). “Performance comparison of heterogeneous classifiers for detection of Parkinson’s disease using voice disorder (dysphonia),” in 2014 International
Conference on Informatics, Electronics & Vision (ICIEV) (Dhaka), 1–7. doi: 10.1109/ICIEV.2014.6850849
[29] Haq, A. U., Li, J. P., Memon, M. H., khan, J., and Malik, A., Ahmad, T., et al. (2019). Feature selection based on L1-Norm support vector machine and effective recognition system for Parkinson’s disease using voice recordings. IEEE Access 7, 37718–37734. doi: 10.1109/ACCESS.2019.2906350
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
