With the rapid development of technology, computers have got in every area of our lives. One of these areas is the area of health. Medical records are now stored in computer environments and can be accessed if needed. Nowadays, thanks to internet technologies, access to information has been facilitated, and this has brought security problems with malicious users. Using computed tomography images in this study, doctor reviews are hidden in these images. For this purpose, similarity based least significant bit method was used. While hiding the reviews of the doctor, three different messages were concealed with the capacity of hiding, applied directly to the hiding and NOT to the reviews to be hidden. The mean square error and peak signal to noise ratio metrics were used as performance evaluation criteria. As a result, the peak signal to noise ratio value of the lowest message hiding capacity (961 bytes) was found as 71.4408, applied to the original reviews, while the peak signal to noise ratio value was found to be 71.0473, applied to the inverted reviews.