卷积神经网络应用于术前静脉期 CT 图像预测胃胃肠道间质瘤患者的风险类别。

Convolutional neural network applied to preoperative venous-phase CT images predicts risk category in patients with gastric gastrointestinal stromal tumors.

机构信息

Department of Radiology, Tongde Hospital of Zhejiang Province, Hangzhou, Zhejiang, China.

Department of radiology, The First Affiliated Hospital of Zhejiang Chinese Medical University (Zhejiang Provincial Hospital of Chinese Medicine), Hangzhou, Zhejiang, China.

出版信息

BMC Cancer. 2024 Mar 1;24(1):280. doi: 10.1186/s12885-024-11962-y.

DOI:10.1186/s12885-024-11962-y

PMID:38429653

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC10908217/

Abstract

OBJECTIVE

The risk category of gastric gastrointestinal stromal tumors (GISTs) are closely related to the surgical method, the scope of resection, and the need for preoperative chemotherapy. We aimed to develop and validate convolutional neural network (CNN) models based on preoperative venous-phase CT images to predict the risk category of gastric GISTs.

METHOD

A total of 425 patients pathologically diagnosed with gastric GISTs at the authors' medical centers between January 2012 and July 2021 were split into a training set (154, 84, and 59 with very low/low, intermediate, and high-risk, respectively) and a validation set (67, 35, and 26, respectively). Three CNN models were constructed by obtaining the upper and lower 1, 4, and 7 layers of the maximum tumour mask slice based on venous-phase CT Images and models of CNN_layer3, CNN_layer9, and CNN_layer15 established, respectively. The area under the receiver operating characteristics curve (AUROC) and the Obuchowski index were calculated to compare the diagnostic performance of the CNN models.

RESULTS

In the validation set, CNN_layer3, CNN_layer9, and CNN_layer15 had AUROCs of 0.89, 0.90, and 0.90, respectively, for low-risk gastric GISTs; 0.82, 0.83, and 0.83 for intermediate-risk gastric GISTs; and 0.86, 0.86, and 0.85 for high-risk gastric GISTs. In the validation dataset, CNN_layer3 (Obuchowski index, 0.871) provided similar performance than CNN_layer9 and CNN_layer15 (Obuchowski index, 0.875 and 0.873, respectively) in prediction of the gastric GIST risk category (All P >.05).

CONCLUSIONS

The CNN based on preoperative venous-phase CT images showed good performance for predicting the risk category of gastric GISTs.

摘要

目的

胃胃肠道间质瘤（GIST）的风险分类与手术方法、切除范围和术前化疗的需要密切相关。我们旨在开发和验证基于术前静脉期 CT 图像的卷积神经网络（CNN）模型，以预测胃 GIST 的风险分类。

方法

作者所在医疗机构 2012 年 1 月至 2021 年 7 月期间经病理诊断为胃 GIST 的 425 例患者分为训练集（154、84 和 59 例分别为极低/低、中、高危）和验证集（67、35 和 26 例分别为）。通过基于静脉期 CT 图像获得最大肿瘤掩模切片的上下 1、4 和 7 层，并分别建立 CNN_layer3、CNN_layer9 和 CNN_layer15 模型，构建了 3 个 CNN 模型。计算曲线下面积（AUROC）和 Obuchowski 指数以比较 CNN 模型的诊断性能。

结果

在验证集中，CNN_layer3、CNN_layer9 和 CNN_layer15 对低危胃 GIST 的 AUROC 分别为 0.89、0.90 和 0.90；对中危胃 GIST 的 AUROC 分别为 0.82、0.83 和 0.83；高危胃 GIST 的 AUROC 分别为 0.86、0.86 和 0.85。在验证数据集中，CNN_layer3（Obuchowski 指数，0.871）在预测胃 GIST 风险分类方面的表现与 CNN_layer9 和 CNN_layer15 相似（Obuchowski 指数分别为 0.875 和 0.873，均为 P>.05）。