A Bounding Box-Based Radiomic Model for Detecting Occult Peritoneal Metastasis in Advanced Gastric Cancer: A Multicenter Study

Abstract

Background: To develop a radiomics model for bounding box-based region of interest (ROI) annotation for the preoperative diagnosis of occult peritoneal metastasis (OPM) in advanced gastric cancer (AGC) patients. Methods: 599 AGC patients from 3 centers were retrospectively enrolled and were divided into training, validation, and testing cohorts. The ROIs for the largest tumor area, the minimum circumscribed rectangle of the ROIs (BBOX), the nonoverlapping area between the ROIs and BBOX (peritumoral area; PERI) and the smallest rectangle that could completely contain the lesion determined by a radiologist (M_BBOX) were used as inputs to extract radiomic features. Multivariate logistic regression was used to construct a radiomics model to estimate the preoperative probability of OPM in GC patients. Findings: The BBOX radiomics model had the highest area under the receiver operating characteristic curve (AUC) of 0.873 [0.820-0.940], outperforming the other radiomics models. The M_BBOX model was not significantly different from BBOX in the validation cohort (AUC: M_BBOX model 0.871 [0.814-0.940] vs. BBOX model 0.873 [0.820-0.940]; p = 0.937). M_BBOX was selected as the final radiomics model because of its extremely low annotation cost and superior OPM discrimination performance (sensitivity of 85.7% and specificity of 82.8%) over the clinical model, and this radiomics model showed comparable diagnostic efficacy in the testing cohort. Interpretation: The radiomics model based on bounding box annotation achieved higher OPM detection accuracy in AGC patients than the clinical model and had potential generalizability. Funding Information: This work was supported by the 1*3*5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (No. ZY2017304). Declaration of Interests: The authors have no relevant conflicts of interest to disclose. Ethics Approval Statement: This multicenter retrospective study was approved by the institutional review board of each center, and the informed consent requirement was waived.