<p>In recent decades, healthcare intelligent control systems has emerged as a widely utilized platform for medical diagnosis, pathological analysis, and case storage. Despite the convenience offered by digital healthcare, it is crucial for people to prioritize the privacy of medical data in order to ensure the security and reliability of health-care intelligent control systems. To tackle these aforementioned concerns, we present a novel universal medical image encryption algorithm that effectively mitigates domain gaps. More specifically, we mitigate the occurrence of multi-domain distribution drift by minimizing the constraints associated with multi-domain latent codes. Furthermore, we have incorporated domain label discriminators within the network architecture to effectively address the complexity arising from multi-domain. We conducted numerous experiments to validate the efficacy of our proposed method. Our qualitative and quantitative experiments showed that our method could successfully mitigate the generation gap across multiple domains. In addition, to gain a deeper understanding of the role of our proposed components, we incorporated network interpretability research to examine the effectiveness of components.</p>