Cardiac Detection Using YOLO-v5 with Data Preprocessing

Currently, interest in medical-related deep learning is dramatically increasing. Although this interest in deep learning is widely used in other fields, but it is very effective in medical image processing such as CT and MRI, which takes a lot of time for simple medical tests and analysis. In general, in deep learning using such image processing, it is possible to determine which algorithm is the most efficient by collecting data, preprocessing data, and using various models This paper conducted a research on cardiovascular CT images collected from Soonchunhyang University Hospital in Korea, and all of them used data collected for 3 years by professional medical staff. In the case of medical data, the number of data is very limited, so the results can vary greatly depending on how it is processed. Therefore, in this paper, research on an efficient deep learning method was conducted through image data preprocessing using Yolo.