文献速递儿科影像腹部DR

深度学习自动分析对比灌肠评估先天性巨结肠

Deep learning-based automated contrast enema analysis to improve the assessment of Hirschsprung disease.
PubMed 文献 · Eur Radiol · 2026-01-22

回顾性单中心研究纳入2011—2023年221例因疑诊先天性巨结肠或其他适应证接受对比灌肠的儿童,共278次检查。基于预训练DenseNet121开发图像分类模型,以直肠活检为参考标准,部分病例以临床随访为参考。结果显示模型在灌肠水平特异度达93.0%,敏感度72.7%,AUC-ROC与放射科专家接近。

要点

  • 模型灌肠水平特异度93.0%高于专家79.1%,适合辅助排除性判断
  • 图像水平敏感度仅58.5%,正常图像识别可靠但病变检出能力有限
  • AUC-ROC为0.830与专家0.804相当,可用于经验不足场景的辅助诊断

核心结论

  • DNN模型在疑诊先天性巨结肠的对比灌肠解读中特异度高于放射科医生,但差异未达到统计学显著。
  • 观察者间一致性为中等(Cohen´s kappa = 0.475),提示模型可作为诊断支持和标准化的辅助工具。

研究设计与样本

  • 回顾性观察性单中心研究,在一家三级医疗中心开展,纳入2011年1月至2023年12月期间接受对比灌肠的儿科患者。
  • 纳入因疑诊先天性巨结肠或其他临床适应证行对比灌肠的患儿,共221例患者、278次对比灌肠检查。
  • 采用预训练深度神经网络DenseNet121构建分类器,参考标准为直肠活检,未行活检者以临床随访为参考。

关键影像表现/征象

  • 模型在灌肠水平识别先天性巨结肠的敏感度为72.7%,特异度为93.0%,平衡准确率为82.8%。

诊断标准与关键数值

  • 单张图像水平:平衡准确率75.3%,敏感度58.5%,特异度92.1%。
  • 对比灌肠水平:平衡准确率82.8%,敏感度72.7%,特异度93.0%。
  • 模型AUC-ROC为0.830,专家原始报告AUC-ROC为0.804,两者相近。
  • 放射科专家特异度为79.1%,低于模型灌肠水平特异度93.0%,但差异无统计学意义。
  • 模型与专家间观察者一致性中等,Cohen´s kappa = 0.475。

临床应用与阅片要点

  • 模型高特异度特点使其在排除性判断中更具参考价值,可用于经验不足场景的辅助。
  • 中等观察者一致性提示模型可用于边缘病例的辅助诊断和报告标准化。
  • 单张图像敏感度较低,临床应用应以灌肠水平整体判断为主。

局限性

  • 回顾性单中心研究设计,结果外推性受限。
  • 模型与专家特异度差异未达到统计学显著。
  • 部分病例未行活检而以临床随访作为参考标准。

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 compare the radiologic assessment of Hirschsprung disease (HD) based on contrast enema with automated image analysis using a deep neural network (DNN) for image recognition.

MATERIALS AND METHODS: A retrospective observational single-centre study was conducted at a tertiary care hospital, including paediatric patients who underwent contrast enema between January 2011 and December 2023, either for suspected HD or other clinical indications. A classifier based on a pretrained DNN (DenseNet121) was developed to detect HD in contrast enema images. DNN performance was assessed using balanced accuracy, sensitivity, and the area under the receiver-operating characteristic curve (AUC-ROC) and area under the precision-recall curve (AUC-PR) analyses. Rectal biopsy was the reference standard, with clinical follow-up in cases where a biopsy was not performed. The DNN classification performance was compared to historical expert radiologic assessment.

RESULTS: A total of 278 contrast enemas were performed in 221 patients (64.8% male, 35.2% female), mean age of 4.14 years and a median of 2.65 years. DenseNet121 achieved 75.3% balanced accuracy, 58.5% sensitivity, and 92.1% specificity per individual image, improving to 82.8%, 72.7%, and 93.0%, respectively, at the contrast enema level. The model achieved a similar AUC-ROC compared to expert radiologists in their original reports (0.830 vs 0.804), and the interobserver agreement was moderate (Cohen´s kappa = 0.475).

CONCLUSION: The DNN model demonstrated higher specificity than radiologists in the interpretation of contrast enemas in patients with suspected HD. Moderate interobserver agreement underscores the model's potential value as a tool for diagnostic support and standardisation, particularly in settings where access to experienced specialists may be limited or in borderline cases.

KEY POINTS: Question Contrast enema is commonly used to evaluate suspected HD, but its diagnostic accuracy is variable and dependent on the radiologist's expertise. Findings A deep learning model outperformed radiologists in specificity (93.0% vs 79.1%), however, the difference was not statistically significant, and the interobserver agreement was moderate (Cohen´s kappa = 0.475). Clinical relevanceA DNN trained for automated analysis of contrast enema can identify patterns suggestive of HD with performance comparable to conventional radiological assessment, underscoring its value as a tool for diagnostic support in borderline cases or when access to experienced specialists may be limited.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Vargova, Varga, Izquierdo Hernandez, et al. Deep learning-based automated contrast enema analysis to improve the assessment of Hirschsprung disease. Eur Radiol. 2026. PMID: 41572045. doi:10.1007/s00330-025-12267-9

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。