文献速递消化内镜消化道肿瘤与早癌

可解释图学习筛查正常内镜下大肠活检:一项回顾性研究

Screening of normal endoscopic large bowel biopsies with interpretable graph learning: a retrospective study.
PubMed 文献 · Gut · 2023-05-12

研究开发了结合病理医生领域知识的图神经网络IGUANA,对3291例患者6591张内镜下大肠活检全切片图像分类为正常或异常。模型在训练与内部验证中AUC-ROC为0.98(SD=0.004),在三个外部数据集测试中平均AUC-ROC=0.97(SD=0.007)。在99%高灵敏度阈值下可减少约55%的正常切片需病理医生复核。

要点

  • IGUANA在外部验证中平均AUC-ROC=0.97(SD=0.007)、AUC-PR=0.97(SD=0.005)。
  • 99%高灵敏度阈值下,模型可减少约55%的正常切片需病理医生复核。
  • 模型提供热图及与组织学特征关联的数值,实现可解释预测。

核心结论

  • 模型取得持续高准确率,显示其在优化日益稀缺的病理医生资源方面的潜力。
  • 可解释预测可指导病理医生诊断决策并增强其对算法的信心,为未来临床采用铺路。

研究设计与样本

  • 开发结合病理医生领域知识的图神经网络,使用临床驱动的可解释特征,将3291例患者(约54%女性、46%男性)的6591张内镜下大肠活检全切片图像分类为正常或异常(非肿瘤性和肿瘤性)。
  • 一个英国国家医疗服务体系(NHS)中心用于模型训练和内部验证,另外两个NHS中心和一个葡萄牙中心的数据用于外部验证。
  • 模型训练和内部验证在2080例患者的5054张全切片图像上进行。
  • 外部测试在来自三个独立外部数据集的1211例患者的1537张全切片图像上进行。

主要发现

  • 模型训练和内部验证的AUC-ROC为0.98(SD=0.004),AUC-PR为0.98(SD=0.003)。
  • 模型(命名为IGUANA)在三个独立外部数据集测试中表现一致,平均AUC-ROC=0.97(SD=0.007),AUC-PR=0.97(SD=0.005)。
  • 在99%高灵敏度阈值下,所提出的模型可将需病理医生复核的正常切片数量减少约55%。
  • IGUANA以热图形式突出全切片图像中的潜在异常,并提供将模型预测与各种组织学特征关联的数值。

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)。

原文

OBJECTIVE: To develop an interpretable artificial intelligence algorithm to rule out normal large bowel endoscopic biopsies, saving pathologist resources and helping with early diagnosis. DESIGN: A graph neural network was developed incorporating pathologist domain knowledge to classify 6591 whole-slides images (WSIs) of endoscopic large bowel biopsies from 3291 patients (approximately 54% female, 46% male) as normal or abnormal (non-neoplastic and neoplastic) using clinically driven interpretable features. One UK National Health Service (NHS) site was used for model training and internal validation. External validation was conducted on data from two other NHS sites and one Portuguese site. RESULTS: Model training and internal validation were performed on 5054 WSIs of 2080 patients resulting in an area under the curve-receiver operating characteristic (AUC-ROC) of 0.98 (SD=0.004) and AUC-precision-recall (PR) of 0.98 (SD=0.003). The performance of the model, named Interpretable Gland-Graphs using a Neural Aggregator (IGUANA), was consistent in testing over 1537 WSIs of 1211 patients from three independent external datasets with mean AUC-ROC=0.97 (SD=0.007) and AUC-PR=0.97 (SD=0.005). At a high sensitivity threshold of 99%, the proposed model can reduce the number of normal slides to be reviewed by a pathologist by approximately 55%. IGUANA also provides an explainable output highlighting potential abnormalities in a WSI in the form of a heatmap as well as numerical values associating the model prediction with various histological features. CONCLUSION: The model achieved consistently high accuracy showing its potential in optimising increasingly scarce pathologist resources. Explainable predictions can guide pathologists in their diagnostic decision-making and help boost their confidence in the algorithm, paving the way for its future clinical adoption.

来源:Gut,查看原文。版权归原作者所有。

引用

  1. Graham S, Minhas F, Bilal M, et al. Screening of normal endoscopic large bowel biopsies with interpretable graph learning: a retrospective study. Gut. 2023. PMID: 37173125. doi:10.1136/gutjnl-2023-329512

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。