Logo image
Back
Secure Database Sharing in Healthcare: An LLM Based HIPAA Compliant Solution for Data Privacy and Security
Conference proceeding   Peer reviewed

Secure Database Sharing in Healthcare: An LLM Based HIPAA Compliant Solution for Data Privacy and Security

Md Abdul Barek, Md Bajlur Rashid, Kamrul Islam Riad, Md Mostafizur Rahman, Sharmin Yeasmin, Md Jobair Hossain Faruk, Hakki Erhan Sevil, Guillermo A Francia III, Hossain Shahriar, Alfredo Cuzzocrea, …
Proceedings 2025 IEEE International Conference on Big Data (BigData), pp.4217-4226
2025 IEEE International Conference on Big Data (IEEE BigData 2025) (Macau, China, 12/07/2025–12/10/2025)
03/06/2026
DOI: https://doi.org/10.1109/BigData66926.2025.11401910

Metrics

1 Record Views

Abstract

In the healthcare software development domain, ensuring the privacy and security of sensitive patient data is crucial, especially when sharing databases for testing, auditing, debugging, or development purposes. The Health Insurance Portability and Accountability Act (HIPAA) imposes strict regulations on the handling and sharing of Protected Health Information (PHI), making secure database sharing a challenging and labor-intensive task. Traditional data masking methods rely on manual processes. These are not only time-consuming but also prone to errors, particularly when new tables or columns are introduced in the existing database. This paper presents a novel approach that utilizes Large Language Model (LLM) to automatically identify sensitive data fields that need masking or encryption before sharing. A comparative analysis of four different LLMs and three classical MLs are conducted using a custom dataset and among the tested models, a fine-tuned BERT model achieved the highest accuracy. The proposed method is also implemented as a web-based application, providing database administrators with a schema-first and AI-driven recommendations for masking sensitive data and securely sharing databases in compliance with HIPAA. This approach minimizes human error, improves scalability, and strengthens data security, ultimately facilitating secure and compliant collaborative software development in the healthcare sector.

Details

Logo image