Segmentation based encryption method for medical images

The increasing need for telemedicine in healthcare industry created a great necessity to secure the transmitted data among medical centers. Medical image encryption (MIE) is an important technique to achieve security for medical images. Many researchers use advanced encryption standard (AES) to ensure the security of medical images. Applying AES encryption method for medical images directly leads to a long processing time; also it results in obvious background regions, which are considered flaws. In this paper we apply information theory (IT) to identify the two regions of a medical image: the region of interest (ROI) and the region of background (ROB). In order to reduce the processing time needed to protect a medical image using AES with a higher level of security, we propose a hybrid encryption, where AES is applied for ROI and a coding method such as Gold code (GC) is applied for the ROB after improvement. The proposed method has a shorter processing time than applying AES for the whole medical image. In addition, it has better security as seen in the related entropy and correlation calculations.