Current Approaches and Challenges in Medical Image Analysis and Visually Explainable Artificial Intelligence as Future Opportunities

Medical imaging plays a significant role in the disease diagnosis process. The automation techniques used today are helping medical professionals. The present methods’ Black-box structure makes it difficult for them to be implemented in the medical field with reliability. As a solution of the black box nature, an option is to use the XAI method. In this review, a discussion is done on the available popular visual sources of medical diagnosis. Further, the discussion is extended to the current position of machine learning and deep learning in the medical image domain. The performance of deep learning and machine learning approaches in medical image analysis is covered in the paper that is being presented. There is also a thorough examination of the difficulties these methods present for processing medical images in the real-time implementation of artificial intelligence techniques. Present XAI approaches as potential remedies for problems with deep learning approaches. In the paper discussion is on the different techniques that are available for the explanation of the outcome of the artificial intelligence models. In the last the future of the XAI in the medical domain are discussed.