Objectives:The accurate prediction of lymph node metastasis (LNM) and lymphovascular invasion (LVI) is crucial for determining treatment strategies for early gastric cancer (EGC). This study aimed to develop and validate a deep learning-based clinical decision support system (CDSS) to predict LNM including LVI in EGC using real-world data.Methods:A deep learning-based CDSS was developed by integrating endoscopic images, demographic data, biopsy pathology, and CT findings from the data of 2927 patients with EGC across five institutions. We compared a transformer-based model to an image-only (basic convolutional neural network (CNN)) model and a multimodal classification (CNN with random forest) model. Internal testing was conducted on 449 patients from the five institutions, and external validation was performed on 766 patients from two other institutions. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), probability density function, and clinical utility curve.Results:In the training, internal, and external validation cohorts, LNM/LVI was observed in 379 (12.95%), 49 (10.91%), 15 (9.09%), and 41 (6.82%) patients, respectively. The transformer-based model achieved an AUC of 0.9083, sensitivity of 85.71%, and specificity of 90.75%, outperforming the CNN (AUC 0.5937) and CNN with random forest (AUC 0.7548). High sensitivity and specificity were maintained in internal and external validations. The transformer model distinguished 91.8% of patients with LNM in the internal validation dataset, and 94.0% and 89.1% in the two different external datasets.Conclusions:We propose a deep learning-based CDSS for predicting LNM/LVI in EGC by integrating real-world data, potentially guiding treatment strategies in clinical settings.
目的：准确预测早期胃癌（EGC）的淋巴结转移（LNM）和淋巴血管侵犯（LVI）对于制定治疗策略至关重要。本研究旨在利用真实世界数据，开发并验证一种基于深度学习的临床决策支持系统（CDSS），用于预测早期胃癌的淋巴结转移（包括淋巴血管侵犯）。 方法：通过整合来自五家机构的2927例早期胃癌患者的内镜图像、人口统计学数据、活检病理学及CT影像数据，开发了一种基于深度学习的临床决策支持系统。我们将基于Transformer的模型与仅使用图像的模型（基础卷积神经网络（CNN））以及多模态分类模型（CNN结合随机森林）进行了比较。内部测试在来自五家机构的449例患者中进行，外部验证则在来自另外两家机构的766例患者中进行。模型性能通过受试者工作特征曲线下面积（AUC）、概率密度函数和临床效用曲线进行评估。 结果：在训练集、内部验证集和两个外部验证集中，分别有379例（12.95%）、49例（10.91%）、15例（9.09%）和41例（6.82%）患者观察到淋巴结转移/淋巴血管侵犯。基于Transformer的模型取得了0.9083的AUC值、85.71%的敏感性和90.75%的特异性，其性能优于CNN模型（AUC 0.5937）和CNN结合随机森林模型（AUC 0.7548）。该模型在内部和外部验证中均保持了较高的敏感性和特异性。在内部验证数据集中，Transformer模型识别出了91.8%的淋巴结转移患者，在两个不同的外部数据集中，这一比例分别为94.0%和89.1%。 结论：我们提出了一种基于深度学习的临床决策支持系统，通过整合真实世界数据来预测早期胃癌的淋巴结转移/淋巴血管侵犯，有望在临床环境中指导治疗策略的制定。
肿瘤(癌症)患者之家