Delta-Radiomics Approach Using Contrast-Enhanced and Noncontrast-Enhanced Computed Tomography Images for Predicting Distant Metastasis in Patients With Borderline Resectable Pancreatic Carcinoma

Abstract

PURPOSE: To predict distant metastasis (DM) in patients with borderline resectable pancreatic carcinoma using delta-radiomics features calculated from contrast-enhanced computed tomography (CECT) and non-CECT images. METHODS AND MATERIALS: Among 250 patients who underwent radiation therapy at our institution between February 2013 and December 2021, 67 patients were deemed eligible. A total of 11 clinical features and 3906 radiomics features were incorporated. Radiomics features were extracted from CECT and non-CECT images, and the differences between these features were calculated, resulting in delta-radiomics features. The patients were randomly divided into the training (70%) and test (30%) data sets for model development and validation. Predictive models were developed with clinical features (clinical model), radiomics features (radiomics model), and a combination of the abovementioned features (hybrid model) using Fine-Gray regression (FG) and random survival forest (RSF). Optimal hyperparameters were determined using stratified 5-fold cross-validation. Subsequently, the developed models were applied to the remaining test data sets, and the patients were divided into high- or low-risk groups based on their risk scores. Prognostic power was assessed using the concordance index, with 95% CIs obtained through 2000 bootstrapping iterations. Statistical significance between the above groups was assessed using Gray's test. RESULTS: At a median follow-up period of 23.8 months, 47 (70.1%) patients developed DM. The concordance indices of the FG-based clinical, radiomics, and hybrid models were 0.548, 0.603, and 0.623, respectively, in the test data set, whereas those of the RSF-based models were 0.598, 0.680, and 0.727, respectively. The RSF-based model, including delta-radiomics features, significantly divided the cumulative incidence curves into two risk groups (P < .05). The feature map of the gray-level size-zone matrix showed that the difference in feature values between CECT and non-CECT images correlated with the incidence of DM. CONCLUSIONS: Delta-radiomics features obtained from CECT and non-CECT images using RSF successfully predict the incidence of DM in patients with borderline resectable pancreatic carcinoma.