P276 Development and validation of a computed tomography-based radiomic nomogram to predict early-onset surgery risk in patients with Crohn’s disease

Yao, J.(1)*;Zhong, Y.(1);Zhou, J.(2);Zhi, M.(1);

(1)The Sixth Affiliated Hospital of Sun Yat-Sen University, Department of Gastroenterology- Guangdong Provincial Key Laboratory of Colorectal and Pelvic Floor Disease, Guangzhou, China;(2)The Sixth Affiliated Hospital of Sun Yat-Sen University, Department of Radiology, Guangzhou, China;

Background

Identifying patients with aggressive Crohn’s disease (CD) threatened by a high risk of early onset surgery is challenging. We aimed to establish and validate a radiomic nomogram to predict one-year surgery risk after the diagnosis of CD, thereby facilitating therapeutic strategies making.

Methods

Patients with CD who had undergone computed tomography enterography (CTE) examination at diagnosis were recruited and randomly divided into training and test cohorts at a ratio of 7:3. Arterial phase CTE images were obtained. Inflamed segments and mesenteric fat were semiautomatically segmented, followed by feature selection and signature building. A radiomic nomogram was constructed and validated using a multivariate logistic regression algorithm. (Figure 1)

Results

A total of 268 eligible patients were retrospectively included, 69 of whom underwent surgery one-year after diagnosis. A total of 1208 features from inflamed segments and 1100 features from peripheral mesenteric fat were extracted, and reduced to 11 and 16 potential predictors, respectively, to construct two radiomic signatures. By incorporating the radiomics signatures and clinical factors, the radiomic-clinical nomogram showed favorable calibration and discrimination in the training cohort, with an area under the curve (AUC) of 0.957, which was confirmed in the test set (AUC, 0.917). Decision curve analysis and net reclassification improvement index demonstrated the clinical usefulness of the nomogram. (Figure 2)

Conclusion

We successfully established and validated a CTE-based radiomic nomogram with both inflamed segment and mesenteric fat simultaneously evaluated to predict one-year surgery risk in CD patients, which assisted in clinical decision-making and individualized management.