Abstract: Objective: To explore the value of extracellular volume fraction (ECV) measured by dual-energy computed tomography (DECT) in differentiating benign and malignant intraductal papillary mucinous neoplasms (IPMNs). Methods: A total of 57 patients with IPMN confirmed by surgical pathology and underwent DECT examination within 1 month before the operation were retrospectively collected from June 2018 to December 2021. According to pathological results, the patients were divided into benign group(28 cases) and malignant group(29 cases), and the iodine concentration (IC) of pancreatic parenchyma within 1cm around the lesions and abdominal aorta on the same picture in the delayed-phase were measured respectively and the value of DECT-ECV was calculated according to the formula: (ICpancreas/ICabdominal aorta) x (1-Hematocrit). The clinical and imaging data of the two groups were analyzed by univariate analysis and multivariate Logistic regression analysis to determine the independent risk factors, and the imaging model, clinical model and imaging combined clinical model were constructed respectively. The performance of each model was evaluated according to the area under curve (AUC) under the receiver operating characteristic (ROC). Results: Multivariate analysis showed that the DECT-ECV, the elevated level of carbohydrate antigen (CA) 19-9 , and the history of acute pancreatitis (AP) were independent risk factors for differentiating benign and malignant IPMNs. The AUC of imaging model (DECT-ECV), clinical model (the elevated level of CA19-9 and the history of AP), and imaging combined clinical model (DECT-ECV, the elevated level of CA19-9 and the history of AP) in diagnosing malignant IPMN were 0.808, 0.759, and 0.884, respectively. The diagnostic efficacy of imaging combined clinical model was significantly higher than that of clinical model. Conclusion: DECT-ECV of the pancreas parenchyma around the lesions can differentiate benign and malignant IPMN, and could provide a certain reference for clinical treatment.