Hibrit Özellik Seçimi ve Makine Öğrenimi ile Konuşmaya Dayalı Parkinson Tanısının Optimize Edilmesi

Abstract

Parkinson hastalığı (PD), konuşma ve yürüme bozuklukları da dahil olmak üzere bilişsel ve motor işlevlerde bozulmalara neden olan yaygın bir nörolojik bozukluktur. Semptomları tedavi etmek için sıklıkla ilaç kullanılır, ancak bir tedavi henüz bulunamamıştır. Etkili hastalık yönetimi, erken teşhise bağlıdır. Bu çalışma, ses sinyali özelliklerine dayanarak Parkinson hastalığı olan bireyleri sağlıklı kontrollerden ayırt etmek için K-En Yakın Komşular (KNN), Çok Katmanlı Algılayıcı (MLP), Aşırı Gradyan Artırma (XGBoost) ve Naive Bayes (NB) dahil olmak üzere makine öğrenimi yaklaşımlarını kullanmaktadır. Sınıflandırma performansını artırmak için, filtre tabanlı yöntemler (korelasyon analizi, Varyans Analizi [ANOVA]), Özyinelemeli Özellik Eleme (RFE) gibi sarmalayıcı tabanlı yaklaşımlar, Rastgele Orman (RF) ve XGBoost önemini içeren gömülü modeller ve Genetik Algoritma (GA) dahil olmak üzere çoklu özellik seçimi teknikleri uygulandı. Ayrıca, en alakalı özellikleri belirlemek için GA ve XGBoost özellik önemini birleştirerek yeni bir hibrit yöntem önerildi. Modeller, sınıf dengesizliğini gidermek için özellik ölçeklendirme ve RandomOverSampler gibi standart ön işleme teknikleri kullanılarak eğitildi ve test edildi. Deneysel sonuçlar, hibrit özellik seçimi yönteminin sınıflandırma doğruluğunu önemli ölçüde iyileştirdiğini göstermektedir; titiz 10 Katlı Katmanlı Çapraz Doğrulama kullanılarak, önerilen yöntem %92.74’lük (XGBoost) sağlam bir Ortalama Doğruluk ve %91.13’lük (KNN) Geometrik Ortalama elde etmiştir. Bu bulgular, evrimsel ve model tabanlı özellik seçiminin entegrasyonunun Parkinson hastalığının teşhisini önemli ölçüde geliştirebileceğini ve ses tabanlı tıbbi karar destek sistemleri için umut vadeden bir yaklaşım sağlayabileceğini göstermektedir.

Keywords

• Parkinson hastalığı
• konuşma sinyali analizi
• hibrit özellik seçimi
• makine öğrenimi
• ses tabanlı teşhis

Optimizing Speech-Based Parkinson’s Diagnosis with Hybrid Feature Selection and Machine Learning

Abstract

Parkinson’s disease (PD) is a common neurological disorder that causes cognitive and motor impairments, including speech and gait abnormalities. Medication is often used to treat symptoms, but a cure is still elusive. Effective disease management depends on early detection. This study employs ML approaches, including K-Nearest Neighbors (KNN), Multilayer Perceptron (MLP), Extreme Gradient Boosting (XGBoost), and Naive Bayes (NB), to distinguish between individuals with PD and healthy controls based on voice signal characteristics. To enhance classification performance, multiple feature selection techniques were applied, including filter-based methods (correlation analysis, Analysis of Variance [ANOVA]), wrapper-based approaches such as Recursive Feature Elimination (RFE), embedded models including Random Forest (RF) and XGBoost importance, and a Genetic Algorithm (GA). Furthermore, a novel hybrid method was proposed by combining GA and XGBoost feature importance to identify the most relevant features. The models were trained and tested using standard preprocessing techniques such as feature scaling and RandomOverSampler to address class imbalance. Experimental results demonstrate that the hybrid feature selection method significantly improved classification accuracy. Using rigorous 10-Fold Stratified Cross-Validation, the proposed method achieved a robust mean accuracy of 92.74% (XGBoost) and a geometric mean of 91.13% (KNN). These findings suggest that integrating evolutionary and model-driven feature selection can significantly enhance the diagnosis of PD, providing a promising approach for voice-based medical decision support systems.

Keywords

• : Parkinson’s disease
• speech signal analysis
• hybrid feature selection
• machine learning
• voice-based diagnosis

Supporting Institution

The authors declare that no financial support was received for the research, authorship, or publication of this study

Ethical Statement

The authors declare that this study does not require any ethics committee approval or special permission.

References

1. Iyer, A., Di Giovanni, M., Clark, D., Hagenaars, S. P., Ritchie, S. J., Muñoz-Sandoval, D., Zhang, X., Leng, J., Waddell, J., Carlile, S., & Simonyan, K. (2023). A machine learning method to process voice samples for identification of Parkinson’s disease. Scientific Reports, 13(1), 20615. https://doi.org/10.1038/s41598-023-47568-w
2. Zesiewicz, T. A. (2019). Parkinson disease. Continuum, 25(4), 896–918. https://doi.org/10.1212/con.0000000000000764
3. Mathur, A., Dwivedi, R. K., & Rastogi, R. (2024). A survey of machine learning-based approaches for alzheimer’s disease prediction. Educational Administration Theory and Practices. https://doi.org/10.53555/kuey.v30i1.5984
4. Shyamala, K., & Navamani, T. M. (2024). Design of an efficient prediction model for early Parkinson’s disease diagnosis. IEEE Access, 12, 137295–137309. https://doi.org/10.1109/access.2024.3421302
5. Mohapatra, S., Swain, B. K., & Mishra, M. (2025). Early Parkinson’s disease identification via hybrid feature selection from multi-feature subsets and optimized catboost with smote. Systems Science & Control Engineering, 13(1). https://doi.org/10.1080/21642583.2025.2498909
6. Wasif, T., Hossain, M. I. U., & Mahmud, A. (2021). Parkinson disease prediction using feature selection technique in machine learning. In 2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT), IEEE, 1–5. https://doi.org/10.1109/icccnt51525.2021.9580151
7. Srinivasan, S., Ramadass, P., Mathivanan, S. K., Panneer Selvam, K., Shivahare, B. D., & Shah, M. A. (2024). Detection of Parkinson disease using multiclass machine learning approach. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-64004-9
8. Alalayah, K. M., Senan, E. M., Atlam, H. F., Ahmed, I. A., & Shatnawi, H. S. A. (2023). Automatic and early detection of Parkinson’s disease by analyzing acoustic signals using classification algorithms based on recursive feature elimination method. Diagnostics, 13(11), 1924. https://doi.org/10.3390/diagnostics13111924

9. Ali, A. M., Salim, F., & Saeed, F. (2023). Parkinson’s disease detection using filter feature selection and a genetic algorithm with ensemble learning. Diagnostics, 13(17), 2816. https://doi.org/10.3390/diagnostics13172816
10. Majed, R. J., Al-Heddi, R. M., & Zeki, A. M. (2023). Parkinson’s disease detection using data mining models: A comparative study. In 2023 4th International Conference on Data Analytics for Business and Industry (ICDABI), IEEE, 290–294. https://doi.org/10.1109/icdabi60145.2023.10629302
11. Bunterngchit, C., & Bunterngchit, Y. (2022). A comparative study of machine learning models for Parkinson’s disease detection. In 2022 International Conference on Decision Aid Sciences and Applications (DASA), IEEE, 465–469. https://doi.org/10.1109/dasa54658.2022.9765159
12. Little, M., McSharry, P., Hunter, E., Spielman, J., & Ramig, L. (2008). Suitability of dysphonia measurements for telemonitoring of Parkinson’s disease. Nature Precedings. https://doi.org/10.1038/npre.2008.2298.1
13. Guyon, I., & Elisseeff, A. (2006). An Introduction to Feature Extraction. Springer Berlin Heidelberg, 1–25. https://doi.org/10.1007/978-3-540-35488-8_1
14. Dogra, V., Singh, A., Verma, S., Kavita, Jhanjhi, N. Z., & Talib, M. N. (2021), Understanding of Data Preprocessing for Dimensionality Reduction Using Feature Selection Techniques in Text Classification. Springer Nature Singapore, 455–464. https://doi.org/10.1007/978-981-16-3153-5_48
15. Kuhn, M., & Johnson, K. (2013). Applied Predictive Modeling. Springer New York. https://doi.org/10.1007/978-1-4614-6849-3
16. Montgomery, D. C. (2017). Design and analysis of experiments. John wiley & sons.
17. Albadr, M. A., Tiun, S., Ayob, M., & AL-Dhief, F. (2020). Genetic algorithm based on natural selection theory for optimization problems. Symmetry, 12(11), 1758. https://doi.org/10.3390/sym12111758
18. Sharanyaa, S., Renjith, P. N., & Ramesh, K. (2020). Classification of Parkinson’s disease using speech attributes with parametric and nonparametric machine learning techniques. In 2020 3rd International Conference on Intelligent Sustainable Systems (ICISS), IEEE, 437–442. https://doi.org/10.1109/iciss49785.2020.9316078
19. Brownlee, J. (2020). Imbalanced Classification with Python: Better Metrics, Balance Skewed Classes, Cost-sensitive Learning. Machine Learning Mastery.
20. Yu, T., & Zhu, H. (2020). Hyper-parameter optimization: A review of algorithms and applications.
21. Hossain, E., Hossain, M. F., & Rahaman, M. A. (2019). A color and texture based approach for the detection and classification of plant leaf disease using knn classifier. In 2019 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE, 1–6. https://doi.org/10.1109/ecace.2019.8679247
22. Abirami, S., & Chitra, P. (2020). Chapter fourteen - energy-efficient edge based real-time healthcare support system. In P. Raj & P. Evangeline, editors, The Digital Twin Paradigm for Smarter Systems and Environments: The Industry Use Cases, volume 117 of Advances in Computers, Elsevier, 339–368. https://doi.org/10.1016/bs.adcom.2019.09.007
23. Zoralio˘glu, Y., Gül, M. F., Azizo˘glu, F., Azizo˘glu, G., & Toprak, A. N. (2023). Predicting academic performance of students using machine learning techniques. In 2023 Innovations in Intelligent Systems and Applications Conference (ASYU), IEEE, 1–6. https://doi.org/10.1109/asyu58738.2023.10296648
24. Cover, T., & Hart, P. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21–27. https://doi.org/10.1109/tit.1967.1053964
25. Bishop, C. M., & Nasrabadi, N. M. (2006). Pattern recognition and machine learning. volume 4. Springer.
26. Ramyachitra, D., & Manikandan, P. (2014). Imbalanced dataset classification and solutions: a review. International Journal of Computing and Business Research (IJCBR), 5(4), 1–29.
27. Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD’16, ACM, 785–794. https://doi.org/10.1145/2939672.2939785
28. Natekin, A., & Knoll, A. (2013). Gradient boosting machines, a tutorial. Frontiers in Neurorobotics, 7. https://doi.org/10.3389/fnbot.2013.00021
29. Abdurrahman, G., & Sintawati, M. (2020). Implementation of xgboost for classification of Parkinson’s disease. Journal of Physics: Conference Series, 1538(1), 012024. https://doi.org/10.1088/1742-6596/1538/1/012024
30. Khanum, D. A., Akhtar, A., Raza, S. A., Ansari, M. I., & Shaikh, H. (2022). Parkinson’s disease detection using machine learning algorithms. International Journal for Research in Applied Science and Engineering Technology, 10(10), 786–790. https://doi.org/10.22214/IJRASET.2022.46272
31. Hani, A. A., Abdulwahab, A. A., Nasser, M., Al-Ali, S., & Almohammad, A. (2024). Comparative analysis of state-of-the-art classifiers for Parkinson’s disease diagnosis. Jurnal Ilmiah Ilmu Terapan Universitas Jambi, 8, 409–423. https://doi.org/10.22437/jiituj.v8i2.32771
32. Mittal, C., & Sharma, A. (2022). Parkinson’s disease detection using different machine learning algorithms. International Journal of Scientific and Research Publications, 12(2), 23. https://doi.org/10.29322/ijsrp.12.02.2022.p12205
33. Elen, A., & Avuclu, E. (2020). A comparison of classification methods for diagnosis of Parkinson’s. International Journal of Intelligent Systems and Applications in Engineering, 8(4), 164–170. https://doi.org/10.18201/ijisae.2020466309