АНОНІМІЗАЦІЯ ДАНИХ В ІНТЕЛЕКТУАЛЬНИХ МЕДИЧНИХ СИСТЕМАХ

The importance of digitizing the medical sector and anonymizing medical data for safeguarding patient privacy cannot be overstated. This work explores machine learning methods for anonymizing medical data, focusing on techniques like Generative Adversarial Networks (GANs) and Perturbation Methods. It emphasizes the automated nature of these methods and their ability to maintain data utility while preserving patient confidentiality. Specifically, the application of GANs, such as medGAN, demonstrates their effectiveness in generating synthetic medical data and addressing data imbalance issues while ensuring patient privacy.