We suggest a method for visualizing minor changes in a medical image. Human vision has natural resolution limitations. Monitoring and analysis of minor changes in the medical image off-limits to human vision allows to draw the doctor's attention to possible problems and helps in disease diagnostics. To monitor the changes in the medical image, we used its transformation based on the solution of a special case of the knapsack problem. Transformation of the medical image allows to capture changes in a digital image with an accuracy of one pixel. The transformed medical image is adapted for the perception of human vision. One-pixel changes in the image are well captured by an untrained operator. Testing the method of the transformed medical image on X-ray images of the temporomandibular joint showed the following: the transformed images of fragments of healthy bone tissue on different parts of the jaw have similar shape and approximately the same color scheme. The shape of the transformed image of bone tissue after treatment is similar to the transformed image of healthy bone tissue. The shape and color scheme of the transformed image of diseased bone tissue do not match the shape and color of the transformed image of healthy bone tissue and the transformed image of bone tissue after treatment. Therewith, relatively small transformed image files can be used for preliminary remote diagnostics. The full files shall be transferred only if the doctor detects particular deviations from the norm on the transformed images. The proposed method of the medical image transformation can be used for remote detection of diseases and in other areas of medicine.