SavedAdd your journal

Journals / Balkan Journal of Electrical and Computer Engineering / 2026 / Cilt: 14

Explainable Parkinson’s Disease Prediction from Keystroke Dynamics Using AdaBoost Ensemble Learning

AdaBoost Topluluk Öğrenmesi Kullanılarak Tuş Vuruşu Dinamiklerinden Açıklanabilir Parkinson Hastalığı Tahmini

Ferdi Özbilgin(GİRESUN ÜNİVERSİTESİ, MÜHENDİSLİK FAKÜLTESİ, ELEKTRİK-ELEKTRONİK MÜHENDİSLİĞİ BÖLÜMÜ)0000-0003-4946-7018
Pages
—
Publication Date
2026-09-16
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results obtained indicate that keystroke dynamics can be used as a non-invasive, scalable PD biomarker that does not require a clinical setting.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Parkinson’s disease (PD) is a common, neurodegenerative disorder with a long asymptomatic period before the manifestation of its well-known motor symptoms and is also notoriously difficult to diagnose in its early stages.
📊
Key Findings & Data:The Tappy Keystroke dataset in the PhysioNet dataset is utilized. 33 features from the hold time, flight time and latency were extracted from the samples through statistical analysis.
💡
Conclusion & Impact:The results obtained indicate that keystroke dynamics can be used as a non-invasive, scalable PD biomarker that does not require a clinical setting.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Parkinson hastalığı · tuş vuruşu dinamikleri · AdaBoost

Abstract

Parkinson’s disease (PD) is a common, neurodegenerative disorder with a long asymptomatic period before the manifestation of its well-known motor symptoms and is also notoriously difficult to diagnose in its early stages. The present study is to use machine learning to achieve the diagnosis of Parkinson’s disease based on keystroke dynamics obtained while the users are typing on the keyboard. The Tappy Keystroke dataset in the PhysioNet dataset is utilized. 33 features from the hold time, flight time and latency were extracted from the samples through statistical analysis. The base classifier is the AdaBoost algorithm after combining with Bayesian optimization. The results compared with the Decision Tree, Naive Bayes, K-Nearest Neighbor, Artificial Neural Network, RUSBoost and Bagging all show that the AdaBoost has the highest accuracy of 87.16% and the AUC of 0.9129. It was observed that ensemble-based methods generally showed more balanced and higher performance compared to single classifiers. The decision process of the most successful model was interpreted using the SHAP method; it was determined that the HoldTime and FlightTime variation measures played a decisive role in classification. The results obtained indicate that keystroke dynamics can be used as a non-invasive, scalable PD biomarker that does not require a clinical setting.

Özet

Parkinson hastalığı (PH), iyi bilinen motor semptomlarının ortaya çıkmasından önce uzun bir asemptomatik döneme sahip olan yaygın bir nörodejeneratif hastalıktır ve erken evrelerinde teşhis edilmesi oldukça güçtür. Bu çalışmada, kullanıcıların klavyede yazı yazarken elde edilen tuş vuruşu dinamiklerine dayanarak Parkinson hastalığının teşhisini gerçekleştirmek amacıyla makine öğrenmesi yöntemlerinden yararlanılmıştır. Çalışmada, PhysioNet veri tabanında yer alan Tappy Keystroke veri seti kullanılmıştır. Örneklerden hold time, flight time ve latency değerlerine ilişkin istatistiksel analizler yoluyla toplam 33 özellik çıkarılmıştır. Temel sınıflandırıcı olarak Bayesyen optimizasyon ile birleştirilen AdaBoost algoritması kullanılmıştır. Elde edilen sonuçlar Decision Tree, Naive Bayes, K-Nearest Neighbor, Artificial Neural Network, RUSBoost ve Bagging yöntemleri ile karşılaştırılmış ve AdaBoost algoritmasının %87.16 doğruluk ve 0.9129 AUC değeri ile en yüksek performansı gösterdiği belirlenmiştir. Ayrıca, topluluk tabanlı yöntemlerin genel olarak tekil sınıflandırıcılara kıyasla daha dengeli ve daha yüksek performans sergilediği gözlemlenmiştir. En başarılı modelin karar süreci SHAP yöntemi kullanılarak yorumlanmış ve HoldTime ile FlightTime değişkenlik ölçütlerinin sınıflandırmada belirleyici rol oynadığı tespit edilmiştir. Elde edilen bulgular, tuş vuruşu dinamiklerinin klinik bir ortam gerektirmeyen, invaziv olmayan ve ölçeklenebilir bir Parkinson hastalığı biyobelirteci olarak kullanılabileceğini göstermektedir.

🔖

Cite This Article

Özbilgin, F. (2026). Explainable Parkinson’s Disease Prediction from Keystroke Dynamics Using AdaBoost Ensemble Learning. Balkan Journal of Electrical and Computer Engineering, Cilt: 14. https://doi.org/10.17694/bajece.1905893

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (30)

  1. Jankovic, J. (2008). Parkinson’s disease: clinical features and diagnosis. Journal of neurology, neurosurgery & psychiatry, 79(4), 368-376.
  2. Poewe, W., et al. (2017). Parkinson disease. Nature reviews Disease primers, 3(1), 17013.
  3. Su, D., et al. (2025). Projections for prevalence of Parkinson’s disease and its driving factors in 195 countries and territories to 2050: modelling study of Global Burden of Disease Study 2021. bmj, 388.
  4. Kaya, D., & Soyukibar, T. E. (2022). Parkinson hastalığı ve parkinsonizm. The Journal of Turkish Family Physician, 13(4), 182-192.
  5. Postuma, R. B., et al. (2015). MDS clinical diagnostic criteria for Parkinson's disease. Movement disorders, 30(12), 1591-1601.
  6. Hess, C. W., & Hallett, M. (2017). The phenomenology of Parkinson's disease. Seminars in neurology, 37(02), 109-117.
  7. Meral, M., Ozbilgin, F., & Durmus, F. (2025). Fine-Tuned Machine Learning Classifiers for Diagnosing Parkinson’s Disease Using Vocal Characteristics: A Comparative Analysis. Diagnostics, 15(5), 645.
  8. Meral, M., & Ozbilgin, F. (2025). Parkinson’s disease diagnosis and severity assessment from gait signals via bayesian-optimized deep learning. Diagnostics, 15(16), 2046.
  9. Özbilgin, F., Kurnaz, C., & Aydın, E. (2024). Non-invasive coronary artery disease identification through the iris and bio-demographic health profile features using stacking learning. Image and Vision Computing, 146, 105046.
  10. Senkaya, Y., Kurnaz, C., & Ozbilgin, F. (2025). Enhancing Alzheimer’s Diagnosis with Machine Learning on EEG: A Spectral Feature-Based Comparative Analysis. Diagnostics, 15(17), 2190.
  11. Alfalahi, H., et al. (2022). Diagnostic accuracy of keystroke dynamics as digital biomarkers for fine motor decline in neuropsychiatric disorders: a systematic review and meta-analysis. Scientific reports, 12(1), 7690.
  12. Demir, B., Ulukaya, S., & Erdem, O. (2023). Detection of Parkinson’s disease with keystroke data. Computer Methods in Biomechanics and Biomedical Engineering, 26(13), 1653-1667.
  13. Tripathi, S., Arroyo-Gallego, T., & Giancardo, L. (2022). Keystroke-dynamics for Parkinson's disease signs detection in an at-home uncontrolled population: a new benchmark and method. IEEE Transactions on Biomedical Engineering, 70(1), 182-192.
  14. Roy, S., Roy, U., Sinha, D., & Pal, R. K. (2023). Imbalanced ensemble learning in determining Parkinson’s disease using Keystroke dynamics. Expert Systems with Applications, 217, 119522.
  15. Tripathi, S., Acien, A., Holmes, A. A., Arroyo-Gallego, T., & Giancardo, L. (2024). Generalizing Parkinson’s disease detection using keystroke dynamics: a self-supervised approach. Journal of the American Medical Informatics Association, 31(6), 1239-1246.
  16. Yang, Y., Guo, B., Zhao, K., Liang, Y., Li, N., & Yu, Z. (2024). MELPD-Detector: Multi-level ensemble learning method based on adaptive data augmentation for Parkinson disease detection via free-KD: Y. Yang et al. CCF Transactions on Pervasive Computing and Interaction, 6(2), 182-198.
  17. Adams, W. R. (2017). High-accuracy detection of early Parkinson's Disease using multiple characteristics of finger movement while typing. PloS one, 12(11), e0188226.
  18. Goldberger, A. L., et al. (2000). PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. circulation, 101(23), e215-e220.
  19. Freund, Y., & Schapire, R. E. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences, 55(1), 119-139.
  20. Mienye, I. D., & Sun, Y. (2022). A survey of ensemble learning: Concepts, algorithms, applications, and prospects. Ieee Access, 10, 99129-99149.
  21. Qawqzeh, Y. K., Alourani, A., & Ghwanmeh, S. (2023). An improved breast cancer classification method using an enhanced AdaBoost classifier. International Journal of Advanced Computer Science and Applications, 14(1).
  22. Fawcett, T. (2006). An introduction to ROC analysis. Pattern recognition letters, 27(8), 861-874.
  23. Powers, D. M. (2020). Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation. arXiv preprint arXiv:2010.16061.
  24. Arroyo-Gallego, T., et al. (2018). Detecting motor impairment in early Parkinson’s disease via natural typing interaction with keyboards: validation of the neuroQWERTY approach in an uncontrolled at-home setting. Journal of medical Internet research, 20(3), e89.
  25. Bernardo, L. S., Damaševičius, R., Ling, S. H., de Albuquerque, V. H. C., & Tavares, J. M. R. (2022). Modified SqueezeNet architecture for Parkinson’s disease detection based on keypress data. biomedicines, 10(11), 2746.
  26. Roy, S., Roy, U., Sinha, D., & Pal, R. K. (2025). A novel approach to identify parkinson’s disease and other similar neural stress by analysing keystrokes on modern active devices with ensemble classification. Multimedia Tools and Applications, 84(29), 36163-36202.
  27. Chauhan, R., Jena, M., & Singh, D. (2025). NeuroKey: A Lightweight AI Tool Using Passive Keystroke Dynamics for Parkinson’s Disease Detection and Monitoring. In International Summit on the Global Internet of Things and Edge Computing (pp. 39-53). Springer.
  28. Tat, T., Chen, G., Xu, J., Zhao, X., Fang, Y., & Chen, J. (2025). Diagnosing Parkinson’s disease via behavioral biometrics of keystroke dynamics. Science Advances, 11(14), eadt6631.
  29. Mienye, I. D., & Sun, Y. (2022). A survey of ensemble learning: Concepts, algorithms, applications, and prospects. Ieee Access, 10, 99129-99149.
  30. Roy, S., Roy, U., Sinha, D., & Pal, R. K. (2025). A novel approach to identify parkinson’s disease and other similar neural stress by analysing keystrokes on modern active devices with ensemble classification. Multimedia Tools and Applications, 84(29), 36163-36202.