根据内镜检查结果准确预测早期胃癌（EGC）的病理结果对于决定内镜切除还是手术切除至关重要。本研究旨在开发一种人工智能 (AI) 模型，利用白光内窥镜图像和视频评估 EGC 的综合病理特征。为了训练该模型，我们回顾性收集了接受内窥镜检查的 EGC 患者的 4,336 张图像并前瞻性纳入了 153 个视频或手术切除。使用一组互斥的 260 张图像和 10 个视频对该模型的性能进行了测试，并与 16 名内窥镜医师（9 名专家和 7 名新手）的性能进行了比较。最后，我们使用来自另一机构的 436 张图像和 89 个视频进行了外部验证。经过训练，该模型对未分化组织学的预测准确率达到 89.7%，对粘膜下侵犯的预测准确率达到 88.0%，对淋巴血管侵犯 (LVI) 的预测准确率达到 87.9%，对淋巴血管侵犯 (LVI) 的预测准确率达到 92.7%。淋巴结转移（LNM），使用内窥镜视频。测试中，模型的曲线下面积值对于未分化组织学为0.992，对于粘膜下浸润为0.902，对于LVI为0.706，对于LNM为0.680。此外，该模型在预测未分化组织学（92.7% vs. 71.6%）、粘膜下浸润（87.3% vs. 72.6%）和LNM（87.7% vs. 72.3%）方面显示出明显高于专家的准确度。外部验证显示未分化组织学和粘膜下浸润的准确度分别为 75.6% 和 71.9%。 AI 可以帮助内窥镜医师对 EGC 的分化状态和浸润深度进行高预测。需要进一步研究来改进 LVI 和 LNM 的检测。© 2024。作者。

Accurate prediction of pathologic results for early gastric cancer (EGC) based on endoscopic findings is essential in deciding between endoscopic and surgical resection. This study aimed to develop an artificial intelligence (AI) model to assess comprehensive pathologic characteristics of EGC using white-light endoscopic images and videos.To train the model, we retrospectively collected 4,336 images and prospectively included 153 videos from patients with EGC who underwent endoscopic or surgical resection. The performance of the model was tested and compared to that of 16 endoscopists (nine experts and seven novices) using a mutually exclusive set of 260 images and 10 videos. Finally, we conducted external validation using 436 images and 89 videos from another institution.After training, the model achieved predictive accuracies of 89.7% for undifferentiated histology, 88.0% for submucosal invasion, 87.9% for lymphovascular invasion (LVI), and 92.7% for lymph node metastasis (LNM), using endoscopic videos. The area under the curve values of the model were 0.992 for undifferentiated histology, 0.902 for submucosal invasion, 0.706 for LVI, and 0.680 for LNM in the test. In addition, the model showed significantly higher accuracy than the experts in predicting undifferentiated histology (92.7% vs. 71.6%), submucosal invasion (87.3% vs. 72.6%), and LNM (87.7% vs. 72.3%). The external validation showed accuracies of 75.6% and 71.9% for undifferentiated histology and submucosal invasion, respectively.AI may assist endoscopists with high predictive performance for differentiation status and invasion depth of EGC. Further research is needed to improve the detection of LVI and LNM.© 2024. The Author(s).