Enhancing Security in Medical Image Informatics using Geometrical Attacks

The objective of this study is to provide security to medical images by embedding medical data (electronic patient record; EPR) along with the images to reduce the bandwidth during communication. Reversible watermarking and digital signature, will provide high security. This application is mainly used in telesurgery. Only the authorized medical experts can explore a patient’s image because of the use of the authentication Kerberos. The proposed study is mainly to restrict unauthorized access to patients’ data. Hence medical image authentication may be achieved without biometric recognition such as finger prints, eye stamps, etc. Since the EPR itself contains a patient’s entire history, after the extraction process medical experts can identify the patient and also the disease information. In future we can embed the EPR inside the medical image after encryption to enhance security. To improve the authentication, medical expert biometric information can be embedded inside the image in the future. Experiments were conducted using more than 500 (512 × 512) image archives in various modalities from the National Institutes of Health, USA, and Aycan sample digital images downloaded from the internet. Almost all images with greater than 15,000 bits showed 60.4–78.9 dB PSNR value with less alteration in the restored image because of compression, not of watermarking, and the average (number of pixels change rate) was 98.9%.