A Contrast-Enhanced CT-Based Deep Learning System for Preoperative Prediction of Colorectal Cancer Staging and RAS Mutation

SIMPLE SUMMARY: This study explored the role of CT-based deep learning in detecting colorectal cancer tumor location and preoperatively predicting the stage and RAS gene mutation status of colorectal cancer patients. The deep learning model we built achieved excellent performance. The detection network based on Yolov7 realized the detection and preoperative staging of colorectal cancer with an average mean accuracy of 0.98 in the validation cohort. The vision transformer-based prediction network achieved accurate prediction of preoperative RAS in colorectal cancer patients, achieving an area under the receiver operating characteristic curve (AUC) of 0.9591 and 0.9554 in the test cohort and the validation cohort, respectively. This study also explored the clinical applications of deep learning models. Based on the proposed detection network and prediction network, we built a deep learning system for clinicians who do not understand deep learning. ABSTRACT: Purpose: This study aimed to build a deep learning system using enhanced computed tomography (CT) portal-phase images for predicting colorectal cancer patients’ preoperative staging and RAS gene mutation status. Methods: The contrast-enhanced CT image dataset comprises the CT portal-phase images from a retrospective cohort of 231 colorectal cancer patients. The deep learning system was developed via migration learning for colorectal cancer detection, staging, and RAS gene mutation status prediction. This study used pre-trained Yolov7, vision transformer (VIT), swin transformer (SWT), EfficientNetV2, and ConvNeXt. 4620, and contrast-enhanced CT images and annotated tumor bounding boxes were included in the tumor identification and staging dataset. A total of 19,700 contrast-enhanced CT images comprise the RAS gene mutation status prediction dataset. Results: In the validation cohort, the Yolov7-based detection model detected and staged tumors with a mean accuracy precision (IoU = 0.5) (mAP_0.5) of 0.98. The area under the receiver operating characteristic curve (AUC) in the test set and validation set for the VIT-based prediction model in predicting the mutation status of the RAS genes was 0.9591 and 0.9554, respectively. The detection network and prediction network of the deep learning system demonstrated great performance in explaining contrast-enhanced CT images. Conclusion: In this study, a deep learning system was created based on the foundation of contrast-enhanced CT portal-phase imaging to preoperatively predict the stage and RAS mutation status of colorectal cancer patients. This system will help clinicians choose the best treatment option to increase colorectal cancer patients’ chances of survival and quality of life.