The security of electronic health records (EHRs) is a significant problem in the diverse environment of the healthcare industry. With the ongoing progress of digital transformation, the task of safeguarding the privacy, integrity, and availability of EHRs becomes more intricate. Imaging techniques such as PET, MRI, CT, and X-ray imaging are needed for medical diagnosis. Through the use of these techniques, medical professionals are able to see and evaluate the inside structures, functions, and anomalies of the human body. Retaining, sharing, and manipulating the diagnostic pictures for a variety of purposes, including as segmentation, feature selection, and image denoising, are common practices. Cryptography can secure sensitive medical picture data during storage and transit. Deep learning can totally change medical photo encryption. This project investigates deep learning algorithms for medical picture encryption to improve medical record confidentiality and safety. This work investigates deep learning models in picture encryption, resolution enhancement, detection and classification, encrypted compression, key generation, and end-to-end encryption. In summary, we provide intriguing insights on research difficulties and possible directions for deep learning-based medical image encryption.