Homomorphic Encryption for Secure Computation of White Patch Area in Binarized Medical Images

In specific categories of medical images, pathological abnormalities appear as white patches where the area of the white patches plays a vital role in the diagnosis of the associated disorders. Currently, most medical records are hosted in Cloud Servers for low-cost storage and efficient dissemination. Invariably, the medical images are encrypted and then stored in the Cloud for security and privacy. Moreover, in many cases, homomorphic encryption is adopted to take advantage of the powerful computing capabilities of the Cloud Servers. In such a scenario, a novel homomorphic encryption method that enables the secured computation of the white patch area of a binary image is presented. For this specific task, the target plaintext image is represented in its binary format to demark the white patches. The secured calculation of the white patch area of the specified image in the Cloud server is achieved based on homomorphic matrix multiplications in the cipher domain. The simulated results show that the performance is better compared to its nearest competitor.