Unified Medical Report Management and Prediction System on Blockchain

The increased use of Electronic Health Records (EHRs) has increased access to patient information in medical facilities, but has also raised long-term concerns regarding the security, confidentiality, and partial control of records. In most work environments, medical information is scattered across hospitals, laboratories, and clinics, hindering transparency and making its sharing and analysis extremely difficult. Most existing systems operate on centralised architectures that are not as hard to manage but can be undermined by data breaches, unauthorised access, and single points of failure. To address these shortcomings, this paper presents a management and prediction model for medical reports that integrates blockchain technology, decentralised storage, and machine learning-based analytics. A permissioned consortium blockchain is responsible for metadata, ownership, and access control of large medical files stored off-chain in the InterPlanetary File System (IPFS) to maximise scalability and efficiency. Anonymised and aggregated data have been analysed using machine learning models to enable predictive analysis without exposing sensitive patient data. The proposed system was tested in controlled experimental scenarios using a simulated healthcare dataset. The results demonstrate improved data integrity, clearer control over access, and greater storage efficiency compared to conventional centralised approaches. Although certain scalability, data availability, and real-world application issues remain, the findings demonstrate that the recommended architecture provides a viable and secure foundation for patient-centred healthcare data management and predictive support.