부산대학교 도서관

This study investigates the use of federated learning in healthcare picture analysis with the goal of improving diagnostic precision while safeguarding patient data privacy. A specialized federated learning framework was created, showing considerable gains in precision, privacy protection, as well as computational effectiveness. Sophisticated security measures, such as access limits and encryption, successfully protected private medical picture data. Blockchain technology in addition to the suggested hybrid cloud architecture offered scalable and secure alternatives for healthcare organizations. Decision-makers can take action based on the practical ramifications. Future research ought to concentrate on customizing federated learning to particular imaging modalities, investigating edge computing applications, and evaluating the long-term advantages and difficulties in the field of healthcare.