调查多参数磁共振成像（MRI）协议在胆囊壁增厚（GWT）表征中的诊断性能，包括定量MRI（扩散加权成像（DWI），像素内不一致运动（IVIM），扩散张量成像（DTI）和动态增强（DCE）灌注MRI）和常规MRI。这项前瞻性研究包括于2020年7月至2022年4月之间进行了多参数MRI的GWT连续成年人。两位放射科医师独立评估了MRI。最终诊断基于手术组织病理学。评估MRI参数与恶性GWT的关联。比较定量MRI参数和常规和多参数MRI的诊断性能的曲线下面积（AUC）。计算两位放射科医师之间的观察者间一致性。评估多参数协议结合先进序列的MRI诊断GWT的效能。 对35名GWT患者（平均年龄56岁，23名女性）进行了评估（25例良性和10例恶性）。与恶性GWT显著相关的定量MRI参数为DWI上的明显扩散系数（p = 0.007）和DTI上的平均扩散率（MD）（p = 0.013），IVIM上的灌注分数（f）（p = 0.033），DCE-MRI上的灌注峰值时间（TTP，p = 0.008）和洗脸速率（p = 0.049）。 TTP的AUC最高，为0.790，其次是MD（0.782）和f（0.742）（p = 0.213），用于预测恶性GWT。与常规MRI相比，多参数MRI诊断恶性GWT的敏感性显著更高（90％对80％，p = 0.045）。两位放射科医师阅读的一致性为相当到完美（kappa = 0.639-1），相关性为中度到强（区间类相关系数= 0.5-0.88）。 结论：包括先进序列的多参数协议提高了MRI区分良性和恶性GWT的诊断性能。 •多参数MRI对于诊断恶性GWT有90％的敏感性和88％的特异性，而常规CE-MRI的敏感性为80％，特异性为88％。 •在定量MRI参数中，TTP（灌注MRI）具有最高的AUC值为0.790，其次是MD（0.782）和IVIM-f（0.742）。 •对于大多数定量MRI参数，存在中度到强的协议（ICC = 0.5-0.88）。

To investigate the diagnostic performance of a multiparametric magnetic resonance imaging (MRI) protocol comprising quantitative MRI (diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), diffusion tensor imaging (DTI), and dynamic contrast-enhanced (DCE) perfusion MRI) and conventional MRI in the characterization of gallbladder wall thickening (GWT).This prospective study comprised consecutive adults with GWT who underwent multiparametric MRI between July 2020 and April 2022. Two radiologists evaluated the MRI independently. The final diagnosis was based on surgical histopathology. The association of MRI parameters with malignant GWT was evaluated. The area under the curve (AUC) for the quantitative MRI parameters and diagnostic performance of conventional, and multiparametric MRI were compared. The interobserver agreement between two radiologists was calculated.Thirty-five patients (mean age, 56 years, 23 females) with GWT (25 benign and ten malignant) were evaluated. The quantitative MRI parameters significantly associated with malignant GWT were apparent diffusion coefficient on DWI (p = 0.007) and mean diffusivity (MD) on DTI (p = 0.013), perfusion fraction (f) on IVIM (p = 0.033), time to peak enhancement (TTP, p = 0.008), and wash in rate (p = 0.049) on DCE-MRI. TTP had the highest AUC of 0.790, followed by MD (0.782) and f (0.742) (p = 0.213) for predicting malignant GWT. Multiparametric MRI had significantly higher sensitivity (90% vs. 80%, p = 0.045) than conventional MRI for diagnosing malignant GWT. The two radiologists' reading had substantial to near-perfect agreement (kappa = 0.639-1) and moderate to strong correlation (interclass correlation coefficient = 0.5-0.88).Multiparametric protocol incorporating advanced sequences improved the diagnostic performance of MRI for differentiating benign and malignant GWT.• Multiparametric MRI had 90% sensitivity and 88% specificity for diagnosing malignant GWT, compared to 80% sensitivity and 88% specificity for conventional CE-MRI. • Among the quantitative MRI parameters, TTP (perfusion-MRI) had the highest AUC of 0.790, followed by MD (0.782) and IVIM-f (0.742). • For most quantitative MRI parameters, there was moderate to strong agreement (ICC = 0.5-0.88).© 2023. The Author(s), under exclusive licence to European Society of Radiology.