Application of deep learning in medical image analysis

Deep learning techniques, particularly convolutional networks, have quickly risen to prominence as the preferred approach for interpreting medical pictures. This article highlights nearly 300 contributions to the area, the most of which were published during the past year, and covers the main deep learning principles relevant to medical picture analysis. We look at how deep learning may be used for picture categorization, object recognition, segmentation, registration, and other applications. Studies in the following areas are summarized briefly: neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, and musculoskeletal. We conclude with a review of the present state of the art, a critical evaluation of outstanding problems, and future research prospects. In an era of medical big data, the benefit of deep learning is that important hierarchical connections within the data may be found algorithmically rather than laboriously hand-crafting features. We go through the fundamentals of medical image classification, localization, detection, segmentation, and registration, as well as their applications.We wrap off by talking about research roadblocks, new patterns, and potential future paths.