文献速递腹部泌尿生殖MR

基于深度学习的模拟增强MRI用于直肠癌评估:一项多中心研究

Deep learning-based simulated contrast-enhanced MRI for rectal cancer evaluation: a multicenter study.
PubMed 文献 · Abdom Radiol (NY) · 2025-12-19

本研究纳入三家学术机构514例经病理确诊的直肠癌患者,利用二维生成对抗网络从平扫T1加权图像生成模拟增强MRI。模拟图像与真实图像结构相似指数0.82,肿瘤重叠Dice系数0.86±0.14,肿瘤大小测量ICC 0.76-0.88,98%的合成图像被评为诊断适用;三名放射科医师对真实扫描的EMVI检测AUC(0.702、0.772、0.713)略高于模拟扫描(0.685、0.694、0.661)。作者认为模拟增强MRI在肿瘤大小、强化区域和图像质量方面与真实扫描具有可比性,可能作为诊断和治疗评估的支持工具。

要点

  • 模拟增强MRI与真实增强扫描肿瘤重叠Dice系数0.86±0.14,图像结构相似指数0.82。
  • 肿瘤大小测量ICC为0.76-0.88,长度差异0.11 mm、深度差异0.10 mm。
  • 98%的合成图像被评为诊断适用,但真实扫描EMVI检测AUC略高于模拟扫描。

核心结论

  • 深度学习生成的模拟增强MRI在肿瘤大小、强化区域和图像质量方面与真实扫描具有可比性,提示其可能作为诊断和治疗评估的支持工具。

研究设计与样本

  • 多中心研究,纳入三家学术机构514例经病理确诊的直肠癌患者,使用二维生成对抗网络从平扫T1加权图像生成模拟增强MRI,并以真实增强MRI为参考;其中104对真实与模拟扫描由三名对对比方法设盲的放射科医师评估。

关键影像表现/征象

  • 模拟扫描与真实扫描的结构相似指数为0.82,肿瘤重叠Dice系数为0.86±0.14。
  • 肿瘤大小测量ICC为0.76-0.88,长度差异0.11 mm、深度差异0.10 mm。
  • 三名放射科医师对真实扫描的EMVI检测AUC分别为0.702、0.772、0.713,对模拟扫描分别为0.685、0.694、0.661。
  • 98%的合成扫描被评为诊断适用。

诊断标准与关键数值

  • 肿瘤重叠Dice系数0.86±0.14,结构相似指数0.82。
  • 肿瘤大小测量ICC为0.76-0.88,长度差异0.11 mm,深度差异0.10 mm。
  • 真实扫描EMVI检测AUC为0.702、0.772、0.713(三名放射科医师),模拟扫描为0.685、0.694、0.661。
  • 98%的合成扫描被评为诊断适用。

临床应用与阅片要点

  • 模拟增强MRI与真实扫描在肿瘤大小和强化区域评估上具有可比性,可能作为诊断和治疗评估的支持工具。
  • 真实扫描的EMVI检测AUC(0.702、0.772、0.713)略高于模拟扫描(0.685、0.694、0.661),阅片时需注意模拟扫描对EMVI评估的限度。

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 assess the feasibility and accuracy of using deep learning to generate simulated contrast-enhanced T1-weighted rectal MRI scans from pre-contrast MRI sequences in rectal cancer patients.

METHODS: This study included 514 patients with pathologically confirmed rectal carcinoma who underwent contrast-enhanced MRI at three academic institutions. A two-dimensional generator adversarial network was used to simulate contrast-enhanced MRI scans from pre-contrast T1-weighted images. Quantitative assessments of image similarity, error metrics, and Dice coefficient for tumor overlap were performed. Three radiologists, blinded to the contrast method, evaluated image quality, tumor enhancement, and extramural vascular invasion (EMVI) in 104 paired real and simulated scans. Tumor size correlation was analyzed with intraclass correlation coefficients (ICCs), and modified Bland-Altman plots were used to assess agreement. ROC curves evaluated diagnostic performance of EMVI detection, and quality scores were compared using the McNemar test.

RESULTS: Simulated scans demonstrated high similarity to real scans (structural similarity index: 0.82, Dice coefficient: 0.86 ± 0.14 for tumor overlap). Tumor size measurements correlated strongly (ICC: 0.76-0.88), with minimal differences in length (0.11 mm) and depth (0.10 mm). ROC analysis revealed slightly higher AUC values for real scans (0.702, 0.772, 0.713) compared to simulated scans (0.685, 0.694, 0.661) for three radiologists, respectively. Nearly all synthetic scans (98%) were rated diagnostically suitable.

CONCLUSION: Deep learning-generated simulated contrast-enhanced MRI scans demonstrated comparability to real scans in terms of tumor size, enhancement area, and image quality, suggesting their potential as a supportive tool for diagnosis and treatment evaluation.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Yang P, Tong N, Li Z, et al. Deep learning-based simulated contrast-enhanced MRI for rectal cancer evaluation: a multicenter study. Abdom Radiol (NY). 2025. PMID: 41417085. doi:10.1007/s00261-025-05311-1

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。