文献速递腹部技术与安全PETCT

基于18F-FDG PET的集成深度学习模型预测结直肠癌患者淋巴血管侵犯

18F-FDG PET-based ensemble deep learning model for the prediction of lymphovascular invasion in colorectal cancer patients.
PubMed 文献 · Abdom Radiol (NY) · 2025-11-11

这项回顾性研究纳入177例术前行18F-FDG PET/CT的结直肠癌患者,构建基于肿瘤感兴趣区的2.5D集成深度学习模型预测淋巴血管侵犯。融合模型在内部测试队列中AUC为0.824、准确率0.815、F1分数0.722,表现优于影像组学和临床模型,并显示出良好的临床效用和校准。

要点

  • 基于18F-FDG PET的2.5D集成深度学习模型可用于预测结直肠癌淋巴血管侵犯。
  • 融合模型在内部测试队列AUC为0.824,优于影像组学模型和临床模型。
  • 模型在训练队列和内部测试队列中均表现出良好的校准和临床效用。

核心结论

  • 基于18F-FDG PET图像的2.5D集成深度学习模型在预测结直肠癌淋巴血管侵犯方面表现良好,有望作为结直肠癌患者的精准医学支持工具。

研究设计与样本

  • 回顾性研究,纳入177例术前行18F-FDG PET/CT的结直肠癌患者,分为训练队列和内部测试队列;基于手动勾画的肿瘤感兴趣区确定三种输入(仅肿瘤VOI、向VOI边界外扩展10 mm的近端瘤周、向VOI边界外扩展20 mm的远端瘤周),评估VGG16、Googlenet、ResNet50、DenseNet201和Vision Transformer五种深度学习算法模型,采用支持向量机整合预测性能良好的模型输出建立融合模型,并构建影像组学模型和临床模型进行对比分析;以AUC、准确率、F1分数、参数量和推理时间进行统计评估。

诊断标准与关键数值

  • 近端瘤周-DenseNet201模型:训练队列AUC=0.840、准确率=0.772、F1分数=0.714;内部测试队列AUC=0.738、准确率=0.796、F1分数=0.645;参数量=18.097 M,推理时间=47.500 ms。
  • 原发灶-ResNet50模型:训练队列AUC=0.746、准确率=0.740、F1分数=0.628;内部测试队列AUC=0.733、准确率=0.704、F1分数=0.619;参数量=23.512 M,推理时间=38.200 ms。
  • 融合模型(整合近端瘤周-DenseNet201模型和原发灶-ResNet50模型):训练队列AUC=0.874、准确率=0.821、F1分数=0.766;内部测试队列AUC=0.824、准确率=0.815、F1分数=0.722。
  • 融合模型表现优于影像组学模型和临床模型。

临床应用与阅片要点

  • 基于18F-FDG PET图像的2.5D集成深度学习模型可能作为结直肠癌患者的精准医学支持工具。

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

OBJECTIVE: The aim of this study was to investigate the diagnostic performance of the 2.5-dimensional (2.5D) ensemble deep learning (DL) model based on 18F-fuorodeoxyglucose (FDG) positron emission tomography (PET) images in predicting lymphovascular invasion (LVI) in colorectal cancer (CRC) patients.

METHODS: In this retrospective study, 177 CRC patients who underwent preoperative 18F-FDG PET/computed tomography were enrolled and assigned to the training cohort or the internal test cohort. Three inputs were determined according to the manually-delineated tumor volume of interest (VOI): the PrimaryLesion-2.5D input containing only the tumor VOI, the ProximalPeritumoral-2.5D input extending 10 mm outward from the VOI boundary, and the DistalPeritumoral-2.5D input extending 20 mm outward from the VOI boundary. Five common DL algorithm models, including VGG16, Googlenet, ResNet50, DenseNet201, and Vision Transformer were evaluated. Support vector machine was used to integrate the model outputs with good prediction performance to establish the Fusion model. The Radiomics and Clinical models were constructed for comparative analysis. The performance of the model was statistically analyzed by the area under the curve (AUC), accuracy, F1-score, parameter amount and inference time.

RESULTS: The ProximalPeritumoral-DenseNet201 model (training cohort: AUC = 0.840, accuracy = 0.772, F1-score = 0.714; internal test cohort: AUC = 0.738, accuracy = 0.796, F1-score = 0.645; parameter amount = 18.097 M, inference time = 47.500 ms) and PrimaryLesion-ResNet50 model (training cohort: AUC = 0.746, accuracy = 0.740, F1-score = 0.628; internal test cohort: AUC = 0.733, accuracy = 0.704, F1-score = 0.619; parameter amount = 23.512 M, inference time = 38.200 ms) achieved an optimal balance between performance and computational efficiency. The performance of the Fusion model combined with the ProximalPeritumoral-DenseNet201 model and PrimaryLesion-ResNet50 model was further improved, with an AUC of 0.874, an accuracy of 0.821, and an F1-score of 0.766 in the training cohort. In the internal test cohort, the AUC was 0.824, the accuracy was 0.815, and the F1-score was 0.722. The Fusion model outperformed the Radiomics and Clinical models. Moreover, it showed good clinical utility and calibration.

CONCLUSION: The 2.5D ensemble DL model based on 18F-FDG PET images performed well for the prediction of LVI in CRC, proving its potential as a precision medical support tool for CRC patients.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Zhao H, Lyu Z, Su Y, et al. 18F-FDG PET-based ensemble deep learning model for the prediction of lymphovascular invasion in colorectal cancer patients. Abdom Radiol (NY). 2025. PMID: 41217477. doi:10.1007/s00261-025-05282-3

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。