Medical assistance system on X-ray inspection process with deep learning

In this study, in the process of X-ray examination, the RGB image of the patient site was taken in advance before the X-ray irradiation. Then a Convolutional Neural Network(CNN) model is used to perform image classification on this RGB image. After the classification result is obtained, it is compared with the doctor’s order to see if they are consistent. In the experiment, the data set collected by Taichung Veterans General Hospital was used for training and testing. The data set covers 49 different categories, and the total number of training and testing sheets exceeds 12,000. The goal of this paper is to establish a real-time error detection and early warning assistant system to improve the quality of medical services and enhance patient safety.