Early Detection and Severity Assessment of Parkinson's Disease Using Speech-Based Machine Learning
DOI:
https://doi.org/10.68337/cpsm.v1.i1.2026-003Keywords:
Parkinson's disease, speech signals, machine learning, UPDRS, random forest regression, telemonitoring, severity predictionAbstract
Parkinson's disease (PD) is a neurodegenerative disease characterized by progressive motor disability and early speech disorders. Early diagnosis and regular follow-up allow timely clinical intervention and a better quality of life. This paper introduces a machine learning framework that identifies PD and estimates its severity from acoustic features derived from speech. The data analyzed are from the Parkinson's Telemonitoring dataset of the UCI Machine Learning Repository, which consists of 5,875 voice recordings. Acoustic characteristics such as jitter, shimmer, harmonicity, and nonlinear vocal parameters capture the speech changes associated with PD. Supervised linear regression, support vector regression (SVR), random forest (RF) regression, and gradient boosting (GB) models, together with an RF and GB ensemble, were trained to predict the motor and total Unified Parkinson's Disease Rating Scale (UPDRS) scores, and the predicted scores were then grouped into discrete severity levels. Model performance was analyzed with accuracy, precision, recall, and F1 score. The RF model achieved the best accuracy, 95.49%, which indicates strong predictive ability.
References
[1] Guilherme C. Oliveira, Nemuel D. Pah, Quoc C. Ngo, Arissa Yoshida, Nícolas B. Gomes, João P. Papa, and Dinesh Kumar, "A pilot study for speech assessment to detect the severity of Parkinson's disease: An ensemble approach," Comput. Biol. Med., vol. 185, Art. no. 109565, 2025, doi: 10.1016/j.compbiomed.2024.109565.
[2] Huanqing Xu, Wei Xie, Mingzhen Pang, Ya Li, Luhua Jin, Fangliang Huang, and Xian Shao, "Non-invasive detection of Parkinson's disease based on speech analysis and interpretable machine learning," Front. Aging Neurosci., vol. 17, Art. no. 1586273, 2025, doi: 10.3389/fnagi.2025.1586273.
[3] Tabea Thies, Elisa Mallick, Johannes Tröger, Ebru Baykara, Doris Mücke, and Michael T. Barbe, "Automatic speech analysis combined with machine learning reliably predicts the motor state in people with Parkinson's disease," npj Parkinson's Dis., vol. 11, Art. no. 105, 2025, doi: 10.1038/s41531-025-00959-4.
[4] Daniel Escobar-Grisales, Tomás Arias-Vergara, Cristian David Ríos-Urrego, Elmar Nöth, Adolfo M. García, and Juan Rafael Orozco-Arroyave, "An automatic multimodal approach to analyze linguistic and acoustic cues on Parkinson's disease patients," in Proc. Interspeech 2023, 2023, pp. 1703-1707, doi: 10.21437/Interspeech.2023-2287.
[5] Hadi Sedigh Malekroodi, Nuwan Madusanka, Byeong-il Lee, and Myunggi Yi, "Speech-based Parkinson's detection using pre-trained self-supervised automatic speech recognition (ASR) models and supervised contrastive learning," Bioengineering, vol. 12, no. 7, Art. no. 728, 2025, doi: 10.3390/bioengineering12070728.
[6] Muna Alqam, Abdulnasir Hossen, Salima Al-Abri, Rami Al-Hmouz, and Muhammet Deveci, "Classification of the severity of speech disorders in patients with Parkinson's disease using machine learning techniques," Int. J. Comput. Intell. Syst., vol. 19, no. 1, Art. no. 31, 2026, doi: 10.1007/s44196-025-01109-7.
[7] Farhad Abedinzadeh Torghabeh, Seyyed Abed Hosseini, and Elham Ahmadi Moghadam, "Enhancing Parkinson's disease severity assessment through voice-based wavelet scattering, optimized model selection, and weighted majority voting," Med. Novel Technol. Devices, vol. 20, Art. no. 100266, 2023, doi: 10.1016/j.medntd.2023.100266.
[8] Mohammad Amran Hossain and Francesco Amenta, "Machine learning-based classification of Parkinson's disease patients using speech biomarkers," J. Parkinson's Dis., vol. 14, no. 1, pp. 95-109, 2024, doi: 10.3233/JPD-230002.
[9] Benedetta Perrone, Federica Amato, and Gabriella Olmo, "Voice classification in Parkinson's disease: A deep learning approach using transformers and error rate metrics," Biomed. Signal Process. Control, vol. 113, Art. no. 108954, 2026, doi: 10.1016/j.bspc.2025.108954.
[10] Matthew Shen, Pouria Mortezaagha, and Arya Rahgozar, "Explainable artificial intelligence to diagnose early Parkinson's disease via voice analysis," Sci. Rep., vol. 15, no. 1, Art. no. 11687, 2025, doi: 10.1038/s41598-025-96575-6.
[11] Lorenzo Simone, Mauro Giuseppe Camporeale, Vito Marco Rubino, Vincenzo Gervasi, and Giovanni Dimauro, "Interpretable early detection of Parkinson's disease through speech analysis," in Artificial Intelligence in Medicine (AIME 2025), R. Bellazzi, J. M. Juarez Herrero, L. Sacchi, and B. Zupan, Eds., Lecture Notes in Artificial Intelligence, vol. 15735. Cham, Switzerland: Springer, 2025, pp. 373-378, doi: 10.1007/978-3-031-95841-0_69.
[12] Tariq Adnan, Abdelrahman Abdelkader, Zipei Liu, Ekram Hossain, Sooyong Park, Md Saiful Islam, and Ehsan Hoque, "A novel fusion architecture for detecting Parkinson's disease using semi-supervised speech embeddings," npj Parkinson's Dis., vol. 11, no. 1, Art. no. 176, 2025, doi: 10.1038/s41531-025-00956-7.
[13] Athanasios Sarlas, Alexandros S. Kalafatelis, Georgios Alexandridis, Michail-Alexandros Kourtis, and Panagiotis Trakadas, "Exploring federated learning for speech-based Parkinson's disease detection," in Proc. 18th Int. Conf. Availability, Reliability and Security (ARES 2023), Benevento, Italy, 2023, doi: 10.1145/3600160.3605088.
[14] Wee Shin Lim, Shu-I Chiu, Pei-Ling Peng, Jyh-Shing Roger Jang, Sol-Hee Lee, Chin-Hsien Lin, and Han-Joon Kim, "A cross-language speech model for detection of Parkinson's disease," J. Neural Transm., vol. 132, no. 4, pp. 579-590, 2025, doi: 10.1007/s00702-024-02874-z.
[15] Anna Favaro, Tianyu Cao, Thomas Thebaud, Jesus Villalba, Ankur Butala, Najim Dehak, and Laureano Moro-Velazquez, "Do phonatory features display robustness to characterize Parkinsonian speech across corpora?" in Proc. Interspeech 2023, 2023, pp. 2388-2392, doi: 10.21437/Interspeech.2023-1784.
Downloads
Published
Versions
- 2026-10-01 (2)
- 2026-09-30 (1)
Data Availability Statement
This study used the publicly available Parkinson's Telemonitoring dataset (UCI Machine Learning Repository).Conference Proceedings Volume
Section
Categories
License
Copyright (c) 2026 The Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in Conference Proceedings in Science and Management are made available under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Authors retain copyright in their work and grant the journal first publication rights. Readers are free to share and adapt the published material for any purpose, including commercial use, provided appropriate credit is given to the author, a link to the license is provided, and any changes are indicated.
