Department of Pathology, Beijing Chao-Yang Hospital, Capital Medical University, Beijing, 100020, China.
School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, China.
Gastric Cancer. 2022 Jul;25(4):751-760. doi: 10.1007/s10120-022-01294-w. Epub 2022 Apr 8.
Distinguishing gastric epithelial regeneration change from dysplasia and histopathological diagnosis of dysplasia is subject to interobserver disagreement in endoscopic specimens. In this study, we developed a method to distinguish gastric epithelial regeneration change from dysplasia and further subclassify dysplasia. Meanwhile, optimized the cross-hospital diagnosis using domain adaption (DA).
897 whole slide images (WSIs) of endoscopic specimens from two hospitals were divided into training, internal validation, and external validation cohorts. We developed a deep learning (DL) with DA (DLDA) model to classify gastric dysplasia and epithelial regeneration change into three categories: negative for dysplasia (NFD), low-grade dysplasia (LGD), and high-grade dysplasia (HGD)/intramucosal invasion neoplasia (IMN). The diagnosis based on the DLDA model was compared to 12 pathologists using 100 gastric biopsy cases.
In the internal validation cohort, the diagnostic performance measured by the macro-averaged area under the receiver operating characteristic curve (AUC) was 0.97. In the independent external validation cohort, our DLDA models increased macro-averaged AUC from 0.67 to 0.82. In terms of the NFD and HGD cases, our model's diagnostic sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were significantly higher than junior and senior pathologists. Our model's diagnostic sensitivity, NPV, was higher than specialist pathologists.
We demonstrated that our DLDA model could distinguish gastric epithelial regeneration change from dysplasia and further subclassify dysplasia in endoscopic specimens. Meanwhile, achieved significant improvement of diagnosis cross-hospital.
在胃镜标本中，区分胃上皮再生变化与异型增生以及异型增生的组织病理学诊断容易受到观察者间的分歧。在这项研究中，我们开发了一种方法来区分胃上皮再生变化与异型增生，并进一步对异型增生进行分类。同时，通过域自适应（DA）优化了跨医院的诊断。
将来自两家医院的 897 张胃镜标本全切片图像（WSI）分为训练集、内部验证集和外部验证集。我们开发了一种具有 DA 的深度学习（DL）与 DA（DLDA）模型，将胃异型增生和上皮再生变化分为三类：无异型增生（NFD）、低级别异型增生（LGD）和高级别异型增生（HGD）/黏膜内肿瘤（IMN）。将基于 DLDA 模型的诊断与 12 位病理学家使用 100 例胃活检病例进行比较。
在内部验证集中，通过接收器操作特征曲线（ROC）下的宏观平均面积（macro-averaged AUC）测量的诊断性能为 0.97。在独立的外部验证集中，我们的 DLDA 模型将宏观平均 AUC 从 0.67 提高到 0.82。就 NFD 和 HGD 病例而言，我们模型的诊断敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）明显高于初级和高级病理学家。我们模型的诊断敏感性、NPV 高于专科病理学家。
我们证明了我们的 DLDA 模型可以区分胃上皮再生变化与异型增生，并进一步对胃镜标本中的异型增生进行分类。同时，实现了跨医院诊断的显著改善。
