RELIABILITY-AWARE SCREENING WITH SUBJECT-INDEPENDENT PARKINSON'S DISEASE CLASSIFIERS: UNCERTAINTY, CALIBRATION, SELECTIVE PREDICTION, AND CONFORMAL EVALUATION
AUTHORS:
Mentor
Affiliation
M.Tech Scholar, Department of Computer Science and Engineering,
JNTUK University College of Engineering Kakinada, Andhra Pradesh, India
JNTUK University College of Engineering Kakinada, Andhra Pradesh, India
DOI: 
SHARE THIS ARTICLE
CC BY 4.0 License:
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
DOWNLOAD ARTICLE
REVIEW REPORT
CITE THIS ARTICLE
Abstract
Handwriting and drawing tasks such as spirals, circles, and meanders, together with pen-movement signals, are widely used to build machine learning models for Parkinson's disease detection. Most studies report classification accuracy but say little about whether the predicted probabilities can be trusted for screening decisions on new people. In this work we take previously trained, subject-independent classifiers - three MobileNetV2 CNNs for circle, spiral, and meander drawings and a random forest for pen signals - and treat them as frozen models. Rather than improving accuracy, we evaluate their reliability using four tools: Monte-Carlo dropout and tree-disagreement uncertainty, temperature-scaling calibration measured by the Expected Calibration Error, selective prediction with risk-coverage analysis, and split conformal prediction, all under a strict subject-independent protocol. Reliability varies clearly across modalities, with the signal and meander models showing stronger reliability characteristics than the circle model under the evaluated criteria. Because the held-out test set has only ten subjects, we present the results as a preliminary, screening-oriented reliability assessment rather than a clinical claim.
Keywords
Parkinson's disease; trustworthy machine learning; uncertainty estimation; Monte-Carlo dropout; probability calibration; Expected Calibration Error; selective prediction; conformal prediction; subject-independent evaluation.
Article Information
PUBLICATION DETAILS
Review Type:
Double-Blind Peer Review
License:
CC BY-NC 4.0
Volume/Issue:
Volume 02, Issue 8
Review Rounds:
2
Article Type:
Research Article
Publication Date:
Aug 31 2026
Article Metrics
Article Views
18
PDF Downloads
3
HOW TO CITE
Copy
Kapilavai, B. (2026). Reliability-Aware Screening with Subject-Independent Parkinson's Disease Classifiers: Uncertainty, Calibration, Selective Prediction, and Conformal Evaluation. International Journal of Science, Strategic Management and Technology, 02(8), 1-9. https://doi.org/10.55041/ijsmt.v2i8.091
Kapilavai, Bhargavi. "Reliability-Aware Screening with Subject-Independent Parkinson's Disease Classifiers: Uncertainty, Calibration, Selective Prediction, and Conformal Evaluation." International Journal of Science, Strategic Management and Technology, vol. 02, no. 8, 2026, pp. 1-9. doi:https://doi.org/10.55041/ijsmt.v2i8.091.
Kapilavai, Bhargavi. "Reliability-Aware Screening with Subject-Independent Parkinson's Disease Classifiers: Uncertainty, Calibration, Selective Prediction, and Conformal Evaluation." International Journal of Science, Strategic Management and Technology 02, no. 8 (2026): 1-9. https://doi.org/https://doi.org/10.55041/ijsmt.v2i8.091.
References
[1] C. R. Pereira et al., "A step towards the automated diagnosis of Parkinson's disease: Analyzing handwriting movements," in Proc. IEEE Int. Symp. Computer-Based Medical Systems (CBMS), 2015, pp. 171-176.
[2] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "MobileNetV2: Inverted residuals and linear bottlenecks," in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2018, pp. 4510-4520.
[3] L. Breiman, "Random forests," Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
[4] Y. Gal and Z. Ghahramani, "Dropout as a Bayesian approximation: Representing model uncertainty in deep learning," in Proc. Int. Conf. Machine Learning (ICML), 2016, pp. 1050-1059.
[5] M. P. Naeini, G. F. Cooper, and M. Hauskrecht, "Obtaining well calibrated probabilities using Bayesian binning," in Proc. AAAI Conf. Artificial Intelligence, 2015, pp. 2901-2907.
[6] G. W. Brier, "Verification of forecasts expressed in terms of probability," Monthly Weather Review, vol. 78, no. 1, pp. 1-3, 1950.
[7] C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, "On calibration of modern neural networks," in Proc. Int. Conf. Machine Learning (ICML), 2017, pp. 1321-1330.
[8] R. El-Yaniv and Y. Wiener, "On the foundations of noise-free selective classification," Journal of Machine Learning Research, vol. 11, pp. 1605-1641, 2010.
[9] Y. Geifman and R. El-Yaniv, "Selective classification for deep neural networks," in Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 4878-4887.
[10] G. Shafer and V. Vovk, "A tutorial on conformal prediction," Journal of Machine Learning Research, vol. 9, pp. 371-421, 2008.
[2] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "MobileNetV2: Inverted residuals and linear bottlenecks," in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2018, pp. 4510-4520.
[3] L. Breiman, "Random forests," Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
[4] Y. Gal and Z. Ghahramani, "Dropout as a Bayesian approximation: Representing model uncertainty in deep learning," in Proc. Int. Conf. Machine Learning (ICML), 2016, pp. 1050-1059.
[5] M. P. Naeini, G. F. Cooper, and M. Hauskrecht, "Obtaining well calibrated probabilities using Bayesian binning," in Proc. AAAI Conf. Artificial Intelligence, 2015, pp. 2901-2907.
[6] G. W. Brier, "Verification of forecasts expressed in terms of probability," Monthly Weather Review, vol. 78, no. 1, pp. 1-3, 1950.
[7] C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, "On calibration of modern neural networks," in Proc. Int. Conf. Machine Learning (ICML), 2017, pp. 1321-1330.
[8] R. El-Yaniv and Y. Wiener, "On the foundations of noise-free selective classification," Journal of Machine Learning Research, vol. 11, pp. 1605-1641, 2010.
[9] Y. Geifman and R. El-Yaniv, "Selective classification for deep neural networks," in Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 4878-4887.
[10] G. Shafer and V. Vovk, "A tutorial on conformal prediction," Journal of Machine Learning Research, vol. 9, pp. 371-421, 2008.
Ethics and Compliance
✓ All ethical standards met
This article has undergone plagiarism screening and double-blind peer review. Editorial policies have been followed. Authors retain copyright under CC BY-NC 4.0 license. The research complies with ethical standards and institutional guidelines.
Indexed In
Similar Articles
Study of AI Ecosystem: A Comparison Between FinTech and Healthcare Sector
(2026)
DOI: 10.55041/ijsmt.v2i4.112
AI Alignment Challenges in Large Language Models: Technical Limitations, Risks, and Future Directions
(2026)
DOI: 10.55041/ijsmt.v2i5.353
Design and Cost Analysis of RO Plant for Bullayya College
(2026)
DOI: 10.55041/ijsmt.v2i4.258
Previous Article
