Improving visual quality of CT Images through enhancement of base and detail layers

Medical imaging enables doctors to provide faster and more accurate diagnoses of different conditions. Medical images are generally very dark and frequently exhibit degradations such as poor detail or low contrast, which may impact the accuracy and speed of diagnosis. This paper proposes an effective and efficient technique to enhance CT images. The image is first decomposed into base and detail layers, which are individually enhanced using adaptive gamma correction, and recombined to obtain the resultant image with better details and brightness without added noise and artifacts such as halo. We present a comprehensive study using 51 test images evaluated by 50 human subjects to compare the performance of the proposed method with the existing state of the art. In addition, we use six commonly used objective metrics to score the images produced by the proposed method and seven existing state of the art enhancement methods. The proposed method outperforms the existing techniques in both objective and subjective evaluations and appears as the most effective way of enhancing medical images’ quality without producing artifacts.