文献速递腹部技术与安全DR

超越诊断:基于腹部X线片的深度学习分析在急诊科住院预测中的应用

Beyond diagnosis: deep-learning-based analysis of hospitalization using abdominal radiographs in the emergency department.
PubMed 文献 · Abdom Radiol (NY) · 2025-12-20

本研究回顾性纳入2021年8月至12月1,585例因腹部症状在急诊科接受腹部X线片的成人患者,并另以112例进行外部验证,构建图像模型、临床模型及二者融合的随机森林预测模型以预测住院。内部验证中融合模型AUROC为0.70(95%CI 0.65-0.76),敏感度0.82,F1分数0.75;放射科医师特异度高(0.80-0.95)但敏感度低(0.13-0.52)、F1分数低(0.23-0.55)。外部验证中融合模型AUROC降至0.60,但图像模型敏感度仍高于放射科医师。

要点

  • 融合模型内部验证AUROC 0.70(95%CI 0.65-0.76)、敏感度0.82、F1分数0.75
  • 放射科医师特异度0.80-0.95,但敏感度仅0.13-0.52、F1分数0.23-0.55
  • 深度学习模型敏感度与F1分数明显高于放射科医师,但特异度较低(0.43)

核心结论

  • 深度学习显著提高了腹部X线片筛查住院风险的临床效用,敏感度和F1分数高于放射科医师,但模型尚需进一步改进才能用于临床实施,且作者认为其可能从传统上信息有限的影像检查中提取预测模式。

研究设计与样本

  • 回顾性研究,纳入2021年8月至12月1,585例因腹部症状接受腹部X线片的成人急诊患者,外部验证112例;采用随机森林分类器构建图像模型、临床模型和融合模型,DenseNet201提取图像特征,并纳入非医师人员可获得的早期临床信息。

关键影像表现/征象

  • 全部腹部X线片中11.1%显示异常表现,5.8%对最终诊断具有特异性。
  • 深度学习模型敏感度和F1分数明显高于放射科医师,但特异度较低(0.43),作者认为适合用于筛查。
  • 外部验证中图像模型的敏感度仍高于放射科医师。

诊断标准与关键数值

  • 融合模型内部验证AUROC 0.70(95%CI 0.65-0.76)、敏感度0.82、F1分数0.75。
  • 放射科医师特异度0.80-0.95、敏感度0.13-0.52、F1分数0.23-0.55。
  • 深度学习模型特异度0.43,低于放射科医师。
  • 外部验证中融合模型AUROC降至0.60。

临床应用与阅片要点

  • 作者认为深度学习模型敏感度和F1分数高于放射科医师,但特异度较低(0.43),适合用于筛查。
  • 作者指出这些模型尚需进一步改进才能用于临床实施。

局限性

  • 外部验证中融合模型性能下降,AUROC为0.60。
  • 深度学习模型特异度较低(0.43)。
  • 作者指出这些模型尚需进一步改进才能用于临床实施。

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: Abdominal radiography (AXR) is routinely performed in emergency departments (ED) but has limited clinical utility. Thus, this study developed deep-learning models using AXR to predict hospitalization in ED patients with abdominal symptoms.

METHODS: This retrospective study included 1,585 adult ED patients with abdominal symptoms who underwent AXR between August and December 2021. External validation included 112 patients. Three prediction models were developed using random forest classifiers: an image model, a clinical model, and a fusion model combining both image and clinical features. DenseNet201 extracted image features, and early clinical information obtainable by non-physician staff was incorporated. The performances of the two radiologists were compared with those of a deep-learning model.

RESULTS: The fusion model achieved an area under the receiver operating characteristic curve (AUROC) of 0.70 (95% confidence interval: 0.65-0.76), sensitivity of 0.82, and an F1-score of 0.75 in internal validation. Human readers showed high specificity (0.80-0.95) but low sensitivity (0.13-0.52) and F1-scores (0.23-0.55). Deep-learning models achieved substantially higher sensitivity and F1-scores than human readers, but with lower specificity (0.43), making them suitable for screening. In the external validation, the performance of the fusion model decreased (AUROC, 0.60), but image-based models maintained a higher sensitivity than human readers. Of all the radiographs, 11.1% showed abnormal findings, and 5.8% were specific to the final diagnosis.

CONCLUSION: Deep learning significantly improved the clinical utility of AXR in screening for hospitalization risk, achieving higher sensitivity and F1-scores than radiologists. While these models require further improvement for clinical implementation, they can potentially extract predictive patterns from traditionally limited imaging studies.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Han YE, Cho Y, Park BJ, et al. Beyond diagnosis: deep-learning-based analysis of hospitalization using abdominal radiographs in the emergency department. Abdom Radiol (NY). 2025. PMID: 41420745. doi:10.1007/s00261-025-05347-3

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。