In this study, we examine the effectiveness of deep learning methods in the accurate detection of aortic injury. We used YOLOX and YOLOv8 models as deep learning methods to detect the aortic injury. In the experiments, our findings highlight the successful identification of the site of injury and the application of the “injured” label to CT images of aortic injury. Finally, our study realized high-accuracy detection not only for Contrast-enhanced CT images but also for Plain CT images without the contrast mediums. |
ACCESS THE FULL ARTICLE
No SPIE Account? Create one
Computed tomography
Object detection
Deep learning
Convolutional neural networks
Image processing