VIEWS ON DEEP LEARNING FOR MEDICAL IMAGE DIAGNOSIS

Abstract

Deep learning models are more often used in the medical field as a result of the rapid development of machine learning, graphics processing technologies, and accessibility of medical imaging data. The convolutional neural network (CNN)-based design, adopted by the medical imaging community to assist doctors in identifying the disease, has exacerbated this situation. This research uses a qualitative methodology. The information used in this study, which explores the ideas of deep learning and convolutional neural networks (CNN), taken from publications or papers on artificial intelligent (AI) Convolutional neural networks has been used in recent years for the analysis of medical image data. CNN's development of its machine learning roots is traced in this study. We also provide a brief mathematical description of CNN as well as the pre-processing process required for medical images before inserting them into CNN. Using CNN in many medical domains, including classification, segmentation, detection, and localization, we evaluate relevant research in the field of medical imaging analysis. It can be concluded that CNN's deep learning view of medical imaging is very helpful for medical parties in their work.