Image Caption and Medical Report Generation Based on Deep Learning: a Review and Algorithm Analysis

Algorithms based on deep learning have been widely applied in automatic image captioning and medical report generation. As one of the most important applications in image captioning, medical report generation has been proved to be essential and effective in smart health and diagnosing. Due to the contraction of rapidly increasing need of radiology diagnosis and lack of report writing experience in most medical institutes, it is vital to work out an automatic medical report generation method. However, as the researchers followed the timeline of paper archives of image captioning, it comes out that there still exist a few problems in generating medical reports, which leads to the gap between the algorithm and medical practice. This study will (1) be a concise review of image captioning methods and medical report generation; (2) analyze the reasons of the phenomenon mentioned above with a case study; (3) give a forecast of medical report generation works.