Advanced Artificial Intelligence Encryption Method for Secure Transfer of Medical Images

With the advancement of medical technology, it is now crucial to diagnose illnesses using medical imaging. From one end of the network to the other, medical pictures frequently travel across its branches. High-level security is therefore necessary. Unauthorised usage of the image's data causes issues. One of the most efficient approaches in this area is encryption, which is one of the strategies used to protect the data in the picture. Here, we focus on two key steps: confusion and dispersion. The deep neural network is modified by the amount of weight that has the most impact on the output, whether it be an intermediate result or an elimination output, in addition to an unorganised structure that is undetected utilising deep neural network techniques. The proposed method used colour and greyscale images by segmenting them according to the Region of Interest of the deep neural network algorithm. In order to randomly construct an unpredictable structure that involved the addition or deletion of columns and as well as the rows, the method was then utilised to randomly distribute the pixels inside the designated region and generate random integers. To demonstrate the value of the suggested approach, the suggested algorithm is assessed in a number of ways and contrasted with the current approaches. The model number of changing pixel rate (NPCR) provides a 99.62 value.