Parkinson’s Disease Prediction Assistant Using Webcam-Based Eye Blinking and Finger-Tapping Analysis.

Author

Omkar Pote, Alfaiz Shaikh, Umair Shaikh, Inamul Shaikh, Prof . G.P. Joshi

Abstract

  Parkinson’s disease (PD) is a progressive neurological disorder that can affect movement, coordination, and motor control. Early identification of relevant symptoms may support further clinical assessment. This paper proposes a webcam-based research system that combines eye-blinking analysis with finger-tapping analysis to study behavioral patterns associated with Parkinson’s disease. The existing approach uses computer vision to analyze finger movements, while the proposed extension incorporates facial landmark tracking to estimate eye-blinking patterns. Quantitative features extracted from both activities are intended to be evaluated using Logistic Regression and Random Forest classifiers. Logistic Regression provides a statistical classification approach, while Random Forest combines multiple decision trees to model potentially complex relationships in the extracted features. Both models will be compared using appropriate performance metrics on labelled data. The system aims to provide a low-cost, non-invasive framework for preliminary research. Its reliability must be established through appropriate data validation and further clinical evaluation. It is not intended to replace professional medical diagnosis.

Keywords

Parkinson’s disease, computer vision, eye-blink analysis, finger-tapping analysis, machine learning, Logistic Regression, Random Forest.

Full Text:

Download Paper PDF

References

1 D. L. Guarín, J. K. Wong, N. R. McFarland, and A. Ramirez-Zamora, “Characterizing Disease Progression in Parkinson’s Disease from Videos of the Finger Tapping Test,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 32, pp. 2293–2301, 2024. doi: 10.1109/TNSRE.2024.3416446.
2. C. M. Bishop, Pattern Recognition and Machine Learning. New York, NY, USA: Springer, 2006.
3. L. Breiman, “Random Forests,” Machine Learning, vol. 45, pp. 5–32, 2001.
4. D. W. Hosmer, S. Lemeshow, and R. X. Sturdivant, Applied Logistic Regression, 3rd ed. Hoboken, NJ, USA: Wiley, 2013.
5. F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.
6. C. Lugaresi et al., “MediaPipe: A Framework for Building Perception Pipelines,” arXiv:1906.08172, 2019
7. R. Szeliski, Computer Vision: Algorithms and Applications, 2nd ed. Cham, Switzerland: Springer, 2022.8. “AI Video Analysis in Parkinson’s Disease: A Systematic Review of the Most Accurate Computer Vision Tools for Diagnosis, Symptom Monitoring, and Therapy Management,” PMC, 2025. Read article.
9. “What the trained eye cannot see: Quantitative kinematics and machine learning detect movement deficits in early-stage Parkinson’s disease from videos,” 2024. Read article.
10. “Deep learning of Parkinson’s movement from video, without human-defined measures,” Journal of the Neurological Sciences, vol. 463, article 123089, 2024. doi: 10.1016/j.jns.2024.123089.
11. “The motor prodromes of Parkinson’s disease: From bedside observation to large-scale application,” PMC, 2021. Read article.
12. “Screening for Parkinson’s disease using ‘computer vision’,” PMC, 2025. Read article..

Share your valuable work from Social Media Buttons