Detection of alterations in watermarked medical images using Fast Fourier Transform and Complex-Valued Neural Network

Medical images contain diagnostic information which can be used for early detection of diseases. These images are watermarked in order to proof its integrity; not modified by unauthorized person, and to ascertain the authenticity, that is, ensuring that the image belong to the correct patient and emanates from the correct source. However, the current problem with the watermarking system used for medical images is distortion introduced during the patient data/information embedding. This factor has hindered proper detection and treatment. This paper proposed a distortion free algorithm based on Fast Fourier Transform and Complex Valued Neural Network (FFT-CVNN) that can be used for watermarking medical images. The qualities of the images were evaluated with both pixel and perceptual-based metrics. Results indicate that the host image and the watermarked image were perceptually indistinguishable and the tamper detector was able to detect any form of forgery or tampering in the watermarked image.