EUVIMEDDécouvrir EUVIMED
Studienprotokoll

A Reproducible Protocol for Explainable Autoencoder-Based Anomaly Detection During Secure Healthcare Cloud Database Migration

Priyanka Nalawade, Prashant Kumbharkar, Pankaj Agarkar, Avanish Kumbharkar, Sudan Jha

Journal of Visualized Experiments · 2026-11-13 · (237)

DOI: 10.3791/72501 ↗

At a glance

Studienprotokoll in Journal of Visualized Experiments, published online on 2026-11-13. Written by Priyanka Nalawade and 4 co-authors.

RésuméRéférences bibliographiquesCiteRelated studies

Abstract (original language)

Secure cloud database migration is essential for healthcare and enterprise systems that manage sensitive and distributed data, but migration activities can introduce risks, including credential exposure, unauthorized access, insecure transmission, and data-integrity compromise. This study presents a reproducible protocol for secure cloud database migration that integrates zero-trust security, encrypted communication, temporary access control, centralized observability, autoencoder-based anomaly detection, and explainable artificial intelligence. The protocol establishes a controlled migration workflow comprising dataset preparation, schema validation, encrypted data transfer, continuous telemetry monitoring, anomaly detection, explainable analysis, and post-migration integrity verification. Migration telemetry is partitioned into training, validation, and testing subsets, and the anomaly detection model is evaluated using accuracy, precision, recall, F1-score, and receiver operating characteristic area under the curve. SHAP and LIME are applied to interpret anomalous migration events and identify influential telemetry features. Controlled experiments using simulated distributed cloud migration scenarios demonstrated high anomaly-detection performance, with the proposed configuration achieving 97.9% accuracy, 97.5% precision, 98.4% recall, and 97.9% F1-score under the evaluated configuration. The proposed protocol also reduced credential exposure and incident-detection times while maintaining high data-integrity validation under the evaluated configuration. The protocol provides a reproducible experimental procedure for studying secure, observable, and interpretable cloud database migration and can be adapted to healthcare and enterprise environments subject to their specific privacy, security, and governance requirements.

Cite

APA

Nalawade, P., Kumbharkar, P., Agarkar, P., Kumbharkar, A., & Jha, S. (2026). A Reproducible Protocol for Explainable Autoencoder-Based Anomaly Detection During Secure Healthcare Cloud Database Migration. Journal of Visualized Experiments. https://doi.org/10.3791/72501

Vancouver

Nalawade P, Kumbharkar P, Agarkar P, Kumbharkar A, Jha S. A Reproducible Protocol for Explainable Autoencoder-Based Anomaly Detection During Secure Healthcare Cloud Database Migration. Journal of Visualized Experiments. 2026. doi:10.3791/72501

BibTeX
@article{Nalawade2026,
  author = {Nalawade, Priyanka and Kumbharkar, Prashant and Agarkar, Pankaj and Kumbharkar, Avanish and Jha, Sudan},
  title = {A Reproducible Protocol for Explainable Autoencoder-Based Anomaly Detection During Secure Healthcare Cloud Database Migration},
  journal = {Journal of Visualized Experiments},
  year = {2026},
  number = {237},
  doi = {10.3791/72501}
}
Start for free →