DataSheet_1_Preoperative CT-Based Deep Learning Model for Predicting Risk Stratification in Patients With Gastrointestinal Stromal Tumors.docx

ObjectiveTo develop and evaluate a deep learning model (DLM) for predicting the risk stratification of gastrointestinal stromal tumors (GISTs).
MethodsPreoperative contrast-enhanced CT images of 733 patients with GISTs were retrospectively obtained from two centers between January 2011 and June 2020. The datasets were split into training (n = 241), testing (n = 104), and external validation cohorts (n = 388). A DLM for predicting the risk stratification of GISTs was developed using a convolutional neural network and evaluated in the testing and external validation cohorts. The performance of the DLM was compared with that of radiomics model by using the area under the receiver operating characteristic curves (AUROCs) and the Obuchowski index. The attention area of the DLM was visualized as a heatmap by gradient-weighted class activation mapping.
ResultsIn the testing cohort, the DLM had AUROCs of 0.90 (95% confidence interval [CI]: 0.84, 0.96), 0.80 (95% CI: 0.72, 0.88), and 0.89 (95% CI: 0.83, 0.95) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. In the external validation cohort, the AUROCs of the DLM were 0.87 (95% CI: 0.83, 0.91), 0.64 (95% CI: 0.60, 0.68), and 0.85 (95% CI: 0.81, 0.89) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. The DLM (Obuchowski index: training, 0.84; external validation, 0.79) outperformed the radiomics model (Obuchowski index: training, 0.77; external validation, 0.77) for predicting risk stratification of GISTs. The relevant subregions were successfully highlighted with attention heatmap on the CT images for further clinical review.
ConclusionThe DLM showed good performance for predicting the risk stratification of GISTs using CT images and achieved better performance than that of radiomics model.

研究目的：开发并评估一款深度学习模型（Deep Learning Model, DLM），用于预测胃肠道间质瘤（Gastrointestinal Stromal Tumors, GISTs）的风险分层。
研究方法：回顾性收集2011年1月至2020年6月间，来自两家医疗中心的733例胃肠道间质瘤患者的术前增强CT影像数据。将数据集划分为训练集（n=241）、测试集（n=104）与外部验证集（n=388）。采用卷积神经网络构建用于胃肠道间质瘤风险分层预测的深度学习模型，并在测试集与外部验证集中对其性能进行评估。以受试者工作特征曲线下面积（Area Under Receiver Operating Characteristic Curve, AUROC）与奥丘夫斯基指数（Obuchowski index）比较该深度学习模型与放射组学模型的性能。通过梯度加权类激活映射（Gradient-weighted Class Activation Mapping, Grad-CAM）将深度学习模型的关注区域以热力图形式可视化。
研究结果：在测试集中，该深度学习模型对低恶性、中恶性与高恶性胃肠道间质瘤的受试者工作特征曲线下面积分别为0.90（95%置信区间[CI]：0.84, 0.96）、0.80（95%置信区间[CI]：0.72, 0.88）与0.89（95%置信区间[CI]：0.83, 0.95）。在外部验证集中，其对低恶性、中恶性与高恶性胃肠道间质瘤的受试者工作特征曲线下面积分别为0.87（95%置信区间[CI]：0.83, 0.91）、0.64（95%置信区间[CI]：0.60, 0.68）与0.85（95%置信区间[CI]：0.81, 0.89）。该深度学习模型（奥丘夫斯基指数：训练集0.84；外部验证集0.79）在胃肠道间质瘤风险分层预测任务中性能优于放射组学模型（奥丘夫斯基指数：训练集0.77；外部验证集0.77）。其关注的相关亚区域可通过CT影像上的注意力热力图成功高亮，以供后续临床审核。
研究结论：本研究开发的深度学习模型基于CT影像可有效实现胃肠道间质瘤的风险分层预测，且性能优于放射组学模型。