Deep Learning for Differentiating Benign From Malignant Bile Duct Dilation on MRCP : Development and Prospective Evaluation of an Xception‐Logistic Regression Ensemble Model

作者：Jiong Liu, Lihong Li, Jing Zhang, Chunmei Yang, Xinqiao Huang, Yan Shu, Xiaopeng He, Jian Shu · 发表于：Journal of Magnetic Resonance Imaging · 年份：2025 · DOI：10.1002/jmri.70200 · 被引用次数：1 · 研究领域：Gallbladder and Bile Duct Disorders、Cholangiocarcinoma and Gallbladder Cancer Studies、Pancreatic and Hepatic Oncology Research

BACKGROUND: Accurate identification of benign and malignant bile duct dilatation (BDD) is needed to determine its management plan. Conventional imaging evaluation is subjective, whereas deep learning (DL) offers potential for automated objective assessment. PURPOSE: To construct and evaluate DL models and ensemble strategies based on magnetic resonance cholangiopancreatography (MRCP) images for identifying benign and malignant BDD. STUDY TYPE: Retrospective and prospective. POPULATION: A retrospective cohort (n = 378; median age, 60 years [range: 14, 90]; 194 male) from two institutions and a prospective cohort (n = 60; median age, 62.5 years [range: 15, 86]; 30 male) were included. Retrospective data were randomly stratified split into training, validation, and internal test sets (2:1:1) and an independent external test set. Benign cases were downsampled to balance class distribution. FIELD STRENGTH/SEQUENCE: 3 T MRCP (3D turbo spin echo: VISTA and SPACE). ASSESSMENT: The primary retrospective endpoint was area under the curve (AUC) across DL algorithms and ensembles. Prospectively, the accuracy, sensitivity, and specificity of the model was compared with those of three radiologists. STATISTICAL TESTS: Group comparisons used Mann-Whitney U and Chi-square tests (p < 0.05). Model performance was evaluated using the Hosmer-Lemeshow test, DeLong's test with Bonferroni correction (α = 0.005), and McNemar's test. RESULTS: The Xception model achieved AUCs of 0.816 (95% CI, 0.788-0....