Privacy-Preserving Patient Consent Management Using Federated Learning and Cloud Technology in Digital Health

Protecting patients' personal information while keeping track of their permission is a major challenge in the constantly evolving world of digital health. This paper provides a new approach to patient consent management that uses cloud computing and federated learning (FL) to ensure patient privacy. Training machine learning models on decentralized data is made feasible by FL, which keeps patient information localized and safe. Scalability and reliable data storage are made possible using cloud technologies. To implement this technique, there is a safe system for managing permission, making it easy for patients to approve or reject using their health records. The system uses advanced cryptographic methods and decentralized IDs to safeguard confidentiality and integrity. Healthcare providers may work together to enhance patient care models using FL without gaining direct access to patient data, allowing them to comply with data protection rules like GDPR and HIPAA. This research uses the Patient Consent Management (PCM) database from the BHITS GitHub repository to establish a privacy-preserving FL model for safe and decentralised healthcare data analysis. The proposed methodology achieved a 98.7% accuracy in permission validation, reduced data exposure by 85%, and enhanced processing efficiency by 60%, enabling safe and scalable administration of patient consent in digital health.