Medical Image Enhancement For Disease Diagnosis

Abstract

These days, medical imaging plays a crucial role in cancer diagnosis.  Nevertheless, in order to lower possible dangers during patient picture acquisition, the quality of these medical photographs is usually lowered.  Computer-aided diagnosis systems, which employ algorithms to find unusual aspects in medical images, have seen notable advancements recently.  In addition to ensuring consistency in the interpretation of images and illnesses, this aids radiologists in increasing diagnostic accuracy.   The quality of medical images, the goal data, is a significant factor in determining the performance level that artificial intelligence systems may attain.  Nevertheless, medical photographs have a different pixel value range than regular digital images that AI algorithms typically analyse, and using such data for training without discrimination may lead to subpar algorithm performance.  In this study, we propose a medical image improvement approach that combines standard digital image processing with medical image processing modules.   This scheme's smooth components and strong contrast are intended to enhance medical picture data.    To confirm that this approach is effective in improving the performance of a medical picture segmentation algorithm, we conducted experimental experiments.