Chest X-ray disease diagnosis method based on YOLO model and pseudo color conversion

Chest disease is one of the most common diseases. Chest X-ray is an important method to examine and diagnose chest diseases. The combination of artificial intelligence and X-ray images can appropriately alleviate the shortage of medical resources and the heavy burden on clinicians. In this paper, we bring in the real-time detection model YOLOv4 in the chest disease diagnosis. Chest X-ray is usually gray image with only 256 gray scales, which does not contain enough information for high precision diagnosis. Therefore, we design a pseudo color conversion method for gray X-ray images. Through the conversion, the gray X-ray images with less information turn into colored X-ray images with more details. And then, we use the YOLOv4 to train and detect the colored chest X-ray images. We evaluate our method in the public datasets, the experimental results show that our method can effectively identify and locate chest X-ray diseases.