文献速递腹部介入MR

基于深度学习的磁共振胰胆管成像全自动胆管分割用于胆道手术规划

Fully automatic bile duct segmentation in magnetic resonance cholangiopancreatography for biliary surgery planning using deep learning.
PubMed 文献 · Eur J Radiol · 2025-09-15

研究纳入249例连续接受标准化三维磁共振胰胆管成像(3D-MRCP)扫描的患者,随机分为训练队列208例和测试队列41例,基于nnU-Net框架构建深度学习语义分割模型,并利用10例外中心3D-MRCP扫描数据测试泛化性。内部测试集和外部测试集的平均Dice相似系数分别为0.9403和0.9070,基于自动分割预测的三维模型与真实值结果的相关系数超过0.95。术中吲哚菁绿(ICG)荧光成像和手术情况验证该模型能准确重建胆道标志,作者认为该模型可为复杂胆道手术的术前规划提供指导。

要点

  • 内部测试集41例平均Dice相似系数0.9403,外部测试集10例为0.9070。
  • 自动分割预测三维模型与真实值结果的相关系数超过0.95。
  • 术中ICG荧光成像和手术情况验证模型能准确重建胆道标志。

核心结论

  • 利用深度学习算法框架可训练出对非扩张胆管进行自动且准确三维重建的AI模型,从而为复杂胆道手术的术前规划提供指导。

研究设计与样本

  • 研究为随机分为训练队列(n=208)和测试队列(n=41)的回顾性研究,共纳入249例连续接受标准化3D-MRCP扫描的患者,由两名肝胆外科医生或放射科医生按行业认证程序手动勾画真实值分割,并由两名专家级医生审核用于胆道手术规划,采用nnU-Net框架构建深度学习语义分割模型,并在来自其他中心的10例3D-MRCP扫描数据集上测试泛化性。

诊断标准与关键数值

  • 内部测试集41例和外部测试集10例的平均Dice相似系数分别为0.9403和0.9070;基于自动分割预测的三维模型与真实值结果的相关系数超过0.95;胆道长度95%一致性界限(LoA)为-4.456至4.781,胆道体积95%一致性界限为-3.404至3.650 ml。

临床应用与阅片要点

  • 作者认为,该模型可用于复杂胆道手术的术前规划,术中吲哚菁绿(ICG)荧光成像和手术情况验证了模型能准确重建胆道标志。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVES: To automatically and accurately perform three-dimensional reconstruction of dilated and non-dilated bile ducts based on magnetic resonance cholangiopancreatography (MRCP) data, assisting in the formulation of optimal surgical plans and guiding precise bile duct surgery.

METHODS: A total of 249 consecutive patients who underwent standardized 3D-MRCP scans were randomly divided into a training cohort (n = 208) and a testing cohort (n = 41). Ground truth segmentation was manually delineated by two hepatobiliary surgeons or radiologists following industry certification procedures and reviewed by two expert-level physicians for biliary surgery planning. The deep learning semantic segmentation model was constructed using the nnU-Net framework. Model performance was assessed by comparing model predictions with ground truth segmentation as well as real surgical scenarios. The generalization of the model was tested on a dataset of 10 3D-MRCP scans from other centers, with ground truth segmentation of biliary structures.

RESULTS: The evaluation was performed on 41 internal test sets and 10 external test sets, with mean Dice Similarity Coefficient (DSC) values of respectively 0.9403 and 0.9070. The correlation coefficient between the 3D model based on automatic segmentation predictions and the ground truth results exceeded 0.95. The 95 % limits of agreement (LoA) for biliary tract length ranged from -4.456 to 4.781, and for biliary tract volume ranged from -3.404 to 3.650 ml. Furthermore, the intraoperative Indocyanine green (ICG) fluorescence imaging and operation situation validated that this model can accurately reconstruct biliary landmarks.

CONCLUSION: By leveraging a deep learning algorithmic framework, an AI model can be trained to perform automatic and accurate 3D reconstructions of non-dilated bile ducts, thereby providing guidance for the preoperative planning of complex biliary surgeries.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Tao H, Wang J, Guo K, et al. Fully automatic bile duct segmentation in magnetic resonance cholangiopancreatography for biliary surgery planning using deep learning. Eur J Radiol. 2025. PMID: 40972245. doi:10.1016/j.ejrad.2025.112415

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。