Blockchain and AI based Medical Record Analysis for Bone Fracture on Cloud Environment

Abstract

The widespread adoption of Electronic Medical Records (EMRs) has transformed healthcare by improving the management and sharing of patient information, and thus enhanced patient care. At the same time, this movement toward digital records and sharing of information has resulted in a heightened concern about the privacy and security of this sensitive patient (health) data. The system proposed here addresses these concerns by incorporating Ethereum blockchain technology, which establishes a secure and immutable record of every exchange of patient data. The benefit of blockchain is that it makes it impossible to alter existing patient data; each unit of data exchange is transparent, can be traced, and once established cannot be altered. Using this approach will improve the security and integrity of the healthcare data management process as a whole. In addition, to protect the sensitive patient information in the healthcare records, the project employs the modern encryption protocol XChaCha20 for all exchanges of sensitive patient information. This means that the encrypted records of patients are stored on the InterPlanetary File System (IPFS). Therefore, in the unfortunate event that someone would gain unauthorized access to the relevant data in the system, they will not be able to read the sensitive patient information and privacy will be preserved. The IPFS storage system provides the reliability and security of storing a file in multiple nodes to protect against file loss and file tampering. The IPFS system also implements the ResNet-50 model, a powerful convolutional neural network (CNN) for analyzing medical images. The ResNet-50 deep learning model was trained to predict bone fractures from x-ray images because it could learn how to tie together multiple sophisticated metrics found in the medical data streams. This AI-assisted fracture detection allows for less manual analysis of an image and enables providers to provide faster and better diagnoses. The earlier a bone fracture is identified, the more likely healthcare professionals can intervene and make decisions that can lead to improved care pathways and outcomes for the person. By combining the Ethereum blockchain, XChaCha20 encryption, IPFS storage, and ResNet-50 AI system, the proposed system is a secure, efficient, and automated way to handle healthcare data that is beneficial for providing more secure data. The integration of these systems ultimately promotes a more reliable and efficacious diagnostic approach leading to enhanced patient treatment and care.