Enhancing Security and Privacy in Healthcare with Generative Artificial Intelligence-Based Detection and Mitigation of Data Poisoning Attacks Software

This study investigated an advanced approach to enhancing security and privacy in healthcare by incorporating artificial intelligence (AI)-based strategies to detect and mitigate data poisoning attacks. The proposed method combined unified learning, homomorphic encryption, and autoencoder-based anomaly detection. It ensured that models were trained in diverse places, protected data, and improved model security. Anomaly identification and mitigation and data poisoning resistance were investigated using simulated medical data. Main results. This approach visualized and assessed model performance. This study offers a complete solution to securing medical data and models against new threats.