Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206201,
        author = {Rajan.R and J. Lin Eby Chandra},
        title = {Blockchain-Based Framework for Secure Electronic Health Records With Machine Learning for Disease Risk Prediction},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {2},
        pages = {458-467},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206201},
        abstract = {Medical records in most healthcare settings are scattered across the many hospitals, laboratories, and clinics a person visits over a lifetime. The practical results are duplicated tests, missing context at the point of care, and patients who cannot easily assemble their own history when they need it most. This paper presents MediChain, a medical history management platform that returns ownership of records to the patient while keeping the underlying documents both confidential and independently verifiable. Each record file is encrypted on the server before it leaves the backend and is stored on the InterPlanetary File System (IPFS); only a compact reference, namely a content identifier together with record metadata, is anchored on an Ethereum-compatible smart contract. The contract governs who may write and read a given patient's records: a doctor must first be verified by an administrator and then explicitly granted access by the patient before any record can be added, and the patient can revoke that access at any time. Alongside this integrity layer, a deliberately decoupled machine-learning service estimates disease-risk probability for diabetes, heart disease, chronic kidney disease, and liver disease from structured laboratory values, returning a calibrated risk band rather than a clinical diagnosis. We evaluate the smart contract with a twenty-case unit-test suite that exercises every authorised and unauthorised access path, we report held-out performance for the four risk models (receiver-operating-characteristic area under the curve from 0.76 to 0.95), and we validate the complete record workflow including server-side encryption and cryptographic integrity checking. The architecture is phased so that the core clinical application, the risk-scoring service, and the blockchain layer can each be switched on independently, which keeps a working system demonstrable at every stage of the build.},
        keywords = {Blockchain, Electronic Health Records, InterPlanetary File System (IPFS), Smart Contracts, Role-Based Access Control, Machine Learning, Disease-Risk Prediction, Data Integrity, FastAPI, Solidity.},
        month = {July},
        }