Bio-inspired wavelet algorithm to improve optical contrast of medical images

Medical images have often unfocused zones that present a challenge for interpretation, as medical diagnosis is given through the optical inspection of the affected region. An algorithm that can adjust the contrast has been developed and presented in this paper. We inspired us from predatory birds, which can focus on the sole region of interest and chose wavelets as a useful mathematical tool. To prevent errors, due to under-exposed real-life images (which can mask medical features) we aimed to develop a reliable algorithm: we tried to improve the optical detection of Computed Tomograph - lung images. We combined a Contrast Limited Adaptive Histogram Equalization with a bihrink filter applied in the wavelet domain. Same as the tactics chosen by predatory birds, wavelets are adjustable devices with can zoom on the region we want to focus, adjusting accordingly the mother wavelet and the iterations level. The results obtained after filtering are accurate, the physiological features are better outlined, and have been graphically displayed for interpretation. As a conclusion, the study focuses on making the optical characteristics of the image better and to limit thus the necessity of contrast substances, which can lead to unnecessary complications (especially in elders and children).