Preoperative T staging of rectal cancer based on radiomics of high resolution T2WI
10.13929/j.1003-3289.201902067
- Author:
Gaoxian LI
1
Author Information
1. Department of Medical Imaging, Shanxi Medical University
- Publication Type:Journal Article
- Keywords:
Magnetic resonance imaging;
Neoplasm staging;
Radiomics;
Rectal neoplasms
- From:
Chinese Journal of Medical Imaging Technology
2019;35(8):1224-1228
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the application value of radiomics features extracted from high-resolution T2WI for preoperative T staging of rectal cancer. Methods: Clinical and imaging data of 191 patients with pathologically confirmed rectal cancer were retrospectively analyzed. Whole tumor ROI was delineated on high-resolution T2WI, and 14 imaging features were extracted using LASSO analysis. The patients were divided into early stage (T0-T2) and local advanced stage (T3-T4) according to the pathological results, and then were divided into training group (n=134) and testing group (n=57) to train and test the model. ROC curve and decision curve analysis (DCA) were used to assess the predictability of the model. Results: Tumors in 68 cases were in early stage, while in 123 cases were in the local advanced stage. In training and testing groups, AUC of the predictive model in distinguishing early and local advanced stage of rectal cancer was 0.927 and 0.885, accuracy was 0.851 and 0.842, specificity was 0.917 and 0.700, sensitivity was 0.814 and 0.919, F1 was 0.861 and 0.829, Brier was 0.149 and 0.159, respectively. Conclusion: Using radiomics model constructed from high-resolution T2WI to preoperatively evaluate T stage of rectal cancer is feasible.
