An Optimized Inter Planetary File System Framework Integrating Federated Learning and Blockchain to Bridge Interoperability and Latency Gaps in Electronic Health Record Systems
In the current era, Electronic Health Record (EHR) systems are widely adopted to store and manage patients' medical information in digital form, as they allow doctors and healthcare professionals to view a patient's complete medical information in an instant. The use of EHR makes healthcare faster, more accurate, and safer, and is therefore an important part of the future of digital healthcare. However, it faces many obstacles in terms of seamless integration (interoperability) and low-latency data acquisition, which directly impacts real-time medical decision-making and the quality of patient care. Integrating Blockchain Technology for EHR management with the InterPlanetary File System (IPFS) and federated learning can improve system performance by reducing the high latency of data retrieval, despite challenges like non-Independent and Identically Distributed (IID) data, client drift, and intermittent connectivity across hospital nodes. To address these challenges, we introduced Adaptive Contextual IPFS Retrieval (ACIR) and asynchronous aggregation. We tested our framework in a simulated environment representing 1,000 hospitals, and the results were promising. Data could be retrieved 65% faster, model training finished 38% sooner, and the system's overall performance improved by 42%. Most importantly, we achieved these improvements while maintaining full compliance with HIPAA and GDPR data privacy standards.
@article{dattatraya2025,
author = {Mhaske Varsha Dattatraya and Ashok Kumar P M. and Jadhav Hema Keshav and Devika Veerkumar Mehta},
title = {{An Optimized Inter Planetary File System Framework Integrating Federated Learning and Blockchain to Bridge Interoperability and Latency Gaps in Electronic Health Record Systems}},
journal = {Journal of Trends in Computer Science and Smart Technology},
volume = {7},
number = {3},
pages = {544-567},
year = {2025},
publisher = {IRO Journals},
doi = {10.36548/jtcsst.2025.3.013},
url = {https://doi.org/10.36548/jtcsst.2025.3.013}
}
Copy Citation