EUVIMEDOpdag EUVIMED
Video/Audio

Experimental Protocols for Developing and Validating Explainable AI Models in Secure Healthcare Diagnostics

Sonali Patil, Meghana Lokhande, Shweta Koparde, Sonali Kothari, Sudan Jha

Journal of Visualized Experiments · 2026-10-16 · (236)

DOI: 10.3791/72412 ↗

At a glance

Video/Audio in Journal of Visualized Experiments, published online on 2026-10-16. Written by Sonali Patil and 4 co-authors. Institution: Department of Computer Engineering, Faculty of Science and Technology, Vishwakarma University. Major MEDLINE topics: Artificial Intelligence. Cited 1 times so far (Crossref).

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Artificial Intelligence (AI) has drastically enhanced healthcare diagnostics with automated disease detection, predictive analytics, and intelligent systems for clinical decision support. Nevertheless, the black-box nature of deep learning models hampers interpretability, clinical trust, and regulatory acceptance and raises concerns about data privacy and cybersecurity. This paper presents comprehensive experimental frameworks for developing and evaluating explainable AI models for secure healthcare diagnostics. The suggested model combines responsible data management, multimodal preprocessing, secure deep learning models, explainable artificial intelligence (XAI) methods, and reproducible validation procedures in a single healthcare diagnostic pipeline. Methods of explaining, such as Grad-CAM, SHAP, LIME, and counterfactual analysis, were used to make the models more transparent and understandable for clinicians. At the same time, federated learning and differential privacy methods were used to securely manage highly sensitive patient data. Experimental evaluation has shown 96% classification accuracy, 94% sensitivity, and 95% specificity. These results were consistent in heterogeneous healthcare datasets. Compared to traditional black-box models, the explainability framework not only enhanced the interpretability of the results but also more accurately identified the regions relevant for diagnosis. Moreover, the secure federated learning architecture minimized privacy leakages without sacrificing the predictive performance and remained robust even when exposed to adversarial perturbations. An external evaluation also confirmed the reproducibility and generalizability of the protocols involved. This methodology offers a comprehensive and reliable foundation for deploying AI-powered diagnostics across multiple healthcare domains: radiology, pathology, oncology, cardiology, neurology, and precision medicine.

Subjects

MeSH terms (MEDLINE)

Artificial IntelligenceHumansFederated LearningDeep Learning

Authors and institutions

  • Department of Computer Engineering, Faculty of Science and Technology, Vishwakarma University.
  • Department of Computer Engineering, Pimpri Chinchwad College of Engineering; meghana.lokhande@pccoepune.org.
  • School of Engineering and Technology, DY Patil University, Pune.
  • Symbiosis Institute of Technology, Symbiosis International (Deemed University).
  • Department of Computer Science and Engineering, Kathmandu University; sudan.jha@ku.edu.np.

Cite

APA

Patil, S., Lokhande, M., Koparde, S., Kothari, S., & Jha, S. (2026). Experimental Protocols for Developing and Validating Explainable AI Models in Secure Healthcare Diagnostics. Journal of Visualized Experiments. https://doi.org/10.3791/72412

Vancouver

Patil S, Lokhande M, Koparde S, Kothari S, Jha S. Experimental Protocols for Developing and Validating Explainable AI Models in Secure Healthcare Diagnostics. Journal of Visualized Experiments. 2026. doi:10.3791/72412

BibTeX
@article{Patil2026,
  author = {Patil, Sonali and Lokhande, Meghana and Koparde, Shweta and Kothari, Sonali and Jha, Sudan},
  title = {Experimental Protocols for Developing and Validating Explainable AI Models in Secure Healthcare Diagnostics},
  journal = {Journal of Visualized Experiments},
  year = {2026},
  number = {236},
  doi = {10.3791/72412}
}
Start for free →