文献速递泌尿生殖腹部MR

基于MRI级联深度学习系统对肾脏小实性肿块良恶性分类的临床验证

Construction and clinical validation of a cascaded deep learning system for classification of benign and malignant solid small renal masses based on MRI.
PubMed 文献 · Eur J Radiol · 2026-07-31

本研究回顾性纳入来自三个中心、经病理证实的965例肾脏小实性肿块(SRM)患者,开发基于多参数MRI的级联深度学习系统,实现自动分割与良恶性分类。内部测试集AUC为0.936,外部测试集AUC为0.788。该系统在内部测试集上优于高年资放射科医师,在两个测试集上均优于低年资医师,且在小于3cm肿瘤亚组中结论一致。

要点

  • 级联深度学习系统可实现肾脏小实性肿块自动分割与良恶性分类,外部测试集AUC为0.788。
  • 模型在内部测试集上显著优于高年资医师,在两个测试集上均显著优于低年资医师。
  • 对小于3cm的肾脏小实性肿块,模型仍保持对低年资医师的优势。

核心结论

  • 级联深度学习系统在多个中心数据上表现稳健,可无创且高效地判别肾脏小实性肿块恶性风险,具有作为临床辅助工具的前景。

研究设计与样本

  • 回顾性收集三个中心经病理证实的肾脏小实性肿块患者MRI数据,机构1的888例中712例用于训练、176例作为内部测试集。
  • 机构2和机构3共提供77例作为外部测试集,研究总纳入965例患者。
  • 级联深度学习系统包含自动分割模块与良恶性分类模块,以病理结果为参考标准。

关键影像表现/征象

  • 模型在内部测试集上AUC为0.936,与人工分割标签模型(AUC 0.944)无显著差异(P=0.671)。
  • 模型在外部测试集上AUC为0.788,与人工分割标签模型(AUC 0.832)无显著差异(P=0.629)。

诊断标准与关键数值

  • 使用自动分割标签的最优分类模型,内部测试集AUC为0.936,外部测试集AUC为0.788。
  • 外部测试集上模型与高年资医师表现相当,内部测试集上模型显著优于高年资医师。
  • 两个测试集上模型均显著优于低年资医师,且该结论在小于3cm肿瘤亚组中保持一致。

临床应用与阅片要点

  • 日常阅片时可借助该级联系统进行肾脏小实性肿块的自动定位、分割及良恶性风险提示,辅助报告书写。
  • 阅片时应关注模型对小于3cm肿块的分类结果,其相对低年资医师的优势提示可用于减少漏诊或误判。
  • 该系统的自动分割模块可减少手动勾画病灶的工作量,适合嵌入结构化报告流程。

局限性

  • 外部测试集样本量较小,仅77例,可能影响对模型泛化性评价的稳健性。
  • 研究为回顾性设计,可能存在选择偏倚,且未报告各机构间MRI设备和扫描序列的标准化细节。

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: This study aims to develop a cascaded deep learning (DL) system based on multiparametric MRI to establish an automated pipeline for the segmentation and classification of small renal masses (SRMs).

MATERIALS AND METHODS: A retrospective collection of SRM patients with pathologically confirmed from three institutions was conducted. MRI data from Institution 1 were randomly divided into a training set and an internal test set. Data from other institutions served as the external test set. A cascaded DL system was developed, incorporating automated segmentation and benign-malignant classification. Diagnostic performance was evaluated using receiver operating characteristic analysis and compared against three radiologists of varying experience.

RESULTS: A total of 965 patients with SRM were included. Institution 1 contributed 888 cases, with 712 used for training and 176 as an internal test set; Institutions 2 and 3 provided 77 cases as an external test set. The optimal classification model using automated segmentation labels achieved AUCs of 0.936 and 0.788 on internal and external test sets, respectively. Performance was comparable to models using manual segmentation (internal: 0.936 vs. 0.944, P = 0.671; external: 0.788 vs. 0.832, P = 0.629). On the external test set, the model performed comparably to the senior radiologist, while it significantly outperformed the senior radiologist on the internal test set. The model significantly outperformed the junior radiologist on both test sets. This finding remained consistent in the subgroup of tumors smaller than 3 cm.

CONCLUSION: The cascaded DL system demonstrated robust performance across multiple centers, enabling non-invasive and efficient discrimination of SRM malignancy, showing promise as a clinical support tool.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Cui, Zeng, Chen, et al. Construction and clinical validation of a cascaded deep learning system for classification of benign and malignant solid small renal masses based on MRI. Eur J Radiol. 2026. PMID: 42556272. doi:10.1016/j.ejrad.2026.113127

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。