Abstract
The proposed chaotic visual cryptography approach for medical images allows for secure sharing of medical information in the form of digital images with only the relevant or authorized persons to access the records. We performed the experimentation on the mini-MIAS dataset of mammograms. The effectiveness of the proposed approach is achieved by generation shares that can be stored in a distributed or centralized database without having the risk of exposing the patient's medical condition. Even if the shares are accessed by unauthorized individuals, the original images and the associated condition of a patient are secure. The results show that even superimposing the shares do not reveal any information about the patient's health. It is necessary to decrypt the shares to obtain the original image by using the chaotic key associated with the patient's identity. This renders the original images secure and private only for the patient and concerned doctor with proper access rights.