三维阴性对比CT胰胆管成像上的腹部多器官分割:深度学习方法比较研究
本研究回顾性纳入111例恶性低位胆道梗阻患者双期增强CT,对门静脉期数据进行标注并比较四种深度学习模型在三维阴性对比CT胰胆管成像上的腹部多器官分割性能。Swin-UNETR 3D整体分割性能最优,对肝脏、十二指肠、胰腺、胆道系统的平均DSC分别为96.12%、75.31%、81.00%、88.64%,其三维体素方法在多数指标上优于二维模型。
要点
- Swin-UNETR 3D对胰腺平均DSC为81.00%±6.33%、ASSD为2.07 mm±1.70 mm,对胆道系统平均DSC为88.64%±6.74%。
- 所有模型对肝脏分割准确度高,对十二指肠、胰腺、胆道系统等更小更复杂器官准确度明显较低。
- Swin-UNETR 3D在十二指肠上取得最高平均DSC并较二维模型改善边界定位。
核心结论
- 比较分析显示,三维体素模型(Swin-UNETR 3D)在3D-nCTCP腹部多器官分割的准确性和稳定性上优于二维模型,可减少手动标注时间,并可能有助于更广泛的临床采用。
研究设计与样本
- 回顾性收集111例恶性低位胆道梗阻患者的双期增强CT数据,由一名专家放射科医师处理并标注门静脉期数据,数据集共25,700幅图像,按91例训练/验证、20例测试划分,实现并定量比较TransUNet 2D、nnU-Net 2D、Swin-UNETR 2D、Swin-UNETR 3D四种分割模型,以Dice相似系数(DSC)和平均对称表面距离(ASSD)评价,并以观察者间变异(IOV)作为临床基准。
诊断标准与关键数值
- Swin-UNETR 3D对肝脏平均DSC为96.12%±1.09%、ASSD为4.60 mm±8.25 mm。
- Swin-UNETR 3D对十二指肠平均DSC为75.31%±11.31%、ASSD为4.42 mm±5.84 mm。
- Swin-UNETR 3D对胰腺平均DSC为81.00%±6.33%、ASSD为2.07 mm±1.70 mm。
- Swin-UNETR 3D对胆道系统平均DSC为88.64%±6.74%、ASSD为5.47 mm±12.13 mm。
- 以DSC和ASSD评价,并以观察者间变异(IOV)作为临床基准。
临床应用与阅片要点
- Swin-UNETR 3D在3D-nCTCP上对肝脏、十二指肠、胰腺、胆道系统的分割性能可供术前规划与诊断参考。
- 对十二指肠、胰腺、胆道系统等更小更复杂器官,各模型分割准确度明显低于肝脏。
- Swin-UNETR 3D的三维体素方法在多数指标上优于其二维对应模型,尤其在十二指肠上取得最高平均DSC并改善边界定位。
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: This study aims to automate segmentation of the biliary and pancreatic systems on 3D negative-contrast CT cholangiopancreatography (3D-nCTCP) to improve preoperative planning and diagnosis.
METHODS: We retrospectively collected dual-phase enhanced CT data from 111 patients with malignant low biliary obstruction. Portal phase data were processed and annotated by an expert radiologist. The dataset, comprising 25,700 images, was split into 91 patients for training/validation and 20 patients for testing. Four state-of-the-art segmentation models, namely TransUNet 2D, nnU-Net 2D, Swin-UNETR 2D, and Swin-UNETR 3D, were implemented and quantitatively compared. Model performance was evaluated using the Dice Similarity Coefficient (DSC) and the Average Symmetric Surface Distance (ASSD), with the inter-observer variability (IOV) serving as the clinical benchmark.
RESULTS: Across all models, segmentation performance exhibited high accuracy for the liver, with notably lower accuracy for smaller, more complex organs (duodenum, pancreas, biliary system). The Swin-UNETR 3D model demonstrated superior overall segmentation performance, particularly for challenging organs, with competitive stability. Swin-UNETR 3D achieved a mean DSC of 96.12% ± 1.09% and ASSD of 4.60 mm ± 8.25 mm for the liver, mean DSC of 75.31% ± 11.31% and ASSD of 4.42 mm ± 5.84 mm for the duodenum, mean DSC of 81.00% ± 6.33% and ASSD of 2.07 mm ± 1.70 mm for the pancreas, and mean DSC of 88.64% ± 6.74% and ASSD of 5.47 mm ± 12.13 mm for the biliary system. The 3D volumetric approach in Swin-UNETR 3D outperformed its 2D counterparts (TransUNet 2D, nnU-Net 2D, Swin-UNETR 2D) in most metrics, particularly for the duodenum, where it achieved the highest mean DSC and improved boundary localization compared to 2D models.
CONCLUSION: The comparative analysis demonstrates the superiority of 3D volumetric models (Swin-UNETR 3D) over 2D models for accurate and stable abdominal multi-organ segmentation on 3D-nCTCP, reduces manual annotation time, and may aid broader clinical adoption.
引用
- Li B, Zhang J, Fu H, et al. Abdominal multi-organ segmentation on 3D negative-contrast CT cholangiopancreatography: a comparative study of deep learning methods. Abdom Radiol (NY). 2025. PMID: 41251738. doi:10.1007/s00261-025-05273-4
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。