文献速递泌尿生殖MR

多模态深度学习术前预测前列腺癌切缘阳性:多中心研究

Preoperative prediction of positive surgical margins in prostate cancer using multimodal deep learning model: a multicenter study.
PubMed 文献 · Abdom Radiol (NY) · 2026-07-25

本研究为回顾性多中心研究,纳入 1177 例接受术前 MRI 和根治性前列腺切除术的前列腺癌患者,来自 8 家机构;1022 例用于训练,155 例用于外部验证。采用特征解耦与多模态融合策略构建 MDESM 模型,整合 MRI 影像特征和临床变量。外部验证中 MDESM 的 AUC 为 0.843(95% CI, 0.770-0.911),优于 DESM 和单纯临床模型。

要点

  • MDESM融合MRI与临床,外部验证切缘阳性AUC 0.843(95% CI, 0.770-0.911)。
  • MDESM显著优于DESM和单纯临床模型,且决策曲线分析净获益更高。
  • 模型有潜力用于前列腺癌术前风险分层和手术规划,但需更大样本验证。

核心结论

  • 外部验证队列中 MDESM 预测前列腺癌根治术后切缘阳性的 AUC 为 0.843(95% CI, 0.770-0.911)。
  • MDESM 的预测性能显著优于 DESM 和单纯临床模型,提示 MRI 与临床特征融合可提高预测准确性和泛化能力。

研究设计与样本

  • 回顾性多中心研究,纳入 1177 例接受术前 MRI 和根治性前列腺切除术的前列腺癌患者,来自 8 家机构。
  • 其中 1022 例来自两家机构用于模型训练,155 例来自六家独立中心作为外部验证队列。
  • 参考标准为根治性前列腺切除术后手术切缘状态;采用特征解耦模型 DESM,并整合影像特征与临床变量构建多模态融合模型 MDESM。

诊断标准与关键数值

  • 外部验证队列中 MDESM 的 AUC 为 0.843(95% CI, 0.770-0.911)。
  • 外部验证队列中 DESM 的 AUC 为 0.711(95% CI, 0.607-0.800),与 MDESM 比较 p = 0.003,Z = 2.936。
  • 外部验证队列中单纯临床模型 AUC 为 0.676(95% CI, 0.577-0.770),与 MDESM 比较 p < 0.001,Z = 3.420。
  • 决策曲线分析显示 MDESM 在多个阈值概率范围内净获益最高。

临床应用与阅片要点

  • 对拟行根治性前列腺切除术的患者,可关注 MRI 与临床特征融合模型输出的切缘阳性风险,辅助术前分层。
  • MDESM 的多模态融合策略优于单纯影像或临床模型,提示综合评估可能提升术前预测价值。
  • 模型可作为影像科与泌尿外科沟通手术计划的参考工具,但不应替代术后病理评估。

局限性

  • 外部验证队列样本量仅为 155 例,结果外推需谨慎。
  • 研究为回顾性设计,存在选择偏倚和混杂因素可能。

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

PURPOSE: This study aimed to develop a deep learning model based on magnetic resonance imaging (MRI) and clinical features for predicting PSM risk after radical prostatectomy (RP).

METHODS: This retrospective multicenter study included 1177 prostate cancer patients who underwent preoperative MRI and RP across eight institutions. A total of 1022 patients from two institutions were used for model training, while 155 patients from six independent centers formed the external validation cohort. A feature disentanglement-based deep learning model (DESM) was developed to isolate disease-specific features from hospital-specific variations. A multimodal fusion model (MDESM) was further constructed by integrating the DESM-derived imaging signature with clinical variables to enhance prediction accuracy and generalizability. Gradient-weighted class activation mapping was applied to provide interpretability by highlighting model attention regions.

RESULTS: In the external validation cohort, MDESM achieved an area under the receiver operating characteristic curve of 0.843 (95% CI, 0.770-0.911), significantly outperforming the DESM (0.711, 95% CI, 0.607-0.800, p = 0.003, Z = 2.936) and the clinical-only model (0.676, 95% CI, 0.577-0.770, p < 0.001, Z = 3.420). Decision curve analysis demonstrated the highest net benefit for MDESM across a range of threshold probabilities.

CONCLUSION: The MDESM model, based on a feature disentanglement and multimodal fusion strategy, demonstrates the potential to effectively combine MRI and clinical data to achieve accurate PSM prediction. This approach offers a promising tool for preoperative risk stratification and surgical planning in prostate cancer.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Fu, Bao, Qiao, et al. Preoperative prediction of positive surgical margins in prostate cancer using multimodal deep learning model: a multicenter study. Abdom Radiol (NY). 2026. PMID: 42501097. doi:10.1007/s00261-026-05672-1

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。