Yang Yuhan, Zhou Yin, Zhou Chen, Zhang Xuemei, Ma Xuelei
Department of Pediatric Surgery, West China Hospital, Sichuan University, No. 17 People's South Road, Chengdu, Sichuan, 610041, China.
State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Guoxue Road 37, Chengdu, 610041, China.
J Magn Reson Imaging. 2022 Dec;56(6):1733-1745. doi: 10.1002/jmri.28160. Epub 2022 Mar 18.
MRI acts as a potential resource for exploration and interpretation to identify tumor characterization by advanced computer-aided diagnostic (CAD) methods.
To evaluate and validate the performance of MRI-based CAD models for identifying low-grade and high-grade soft tissue sarcoma (STS) and for investigating survival prognostication.
Retrospective.
A total of 540 patients (295 male/female: 295/245, median age: 42 years) with STSs.
5-T MRI with T WI sequence and fat-suppressed T -weighted (T FS) sequence.
Manual regions of interests (ROIs) were delineated for generation of radiomic features. Automatic segmentation and pretrained convolutional neural networks (CNNs) were performed for deep learning (DL) analysis. The last fully connected layer at the top of CNNs was removed, and the global max pooling was added to transform feature maps to numeric values. Tumor grade was determined on histological specimens.
The support vector machine was adopted as the classifier for all MRI-based models. The DL signature was derived from the DL-MRI model with the highest area under the curve (AUC). The significant clinical variables, tumor location and size, integrated with radiomics and DL signatures were ready for construction of clinical-MRI nomogram to identify tumor grading. The prognostic value of clinical variables and these MRI-based signatures for overall survival (OS) was evaluated via Cox proportional hazard.
The clinical-MRI differentiation nomogram represented an AUC of 0.870 in the training cohort, and an AUC of 0.855, accuracy of 79.01%, sensitivity of 79.03%, and specificity of 78.95% in the validation cohort. The prognostic model showed good performance for OS with 3-year C-index of 0.681 and 0.642 and 5-year C-index of 0.722 and 0.676 in the training and validation cohorts.
MRI-based CAD nomogram represents effective abilities in classification of low-grade and high-grade STSs. The MRI-based prognostic model yields favorable preoperative capacities to identify long-term survivals for STSs.
3 TECHNICAL EFFICACY: Stage 4.
磁共振成像（MRI）可作为一种潜在资源，通过先进的计算机辅助诊断（CAD）方法进行探索和解读，以识别肿瘤特征。
评估和验证基于MRI的CAD模型在识别低级别和高级别软组织肉瘤（STS）以及研究生存预后方面的性能。
回顾性研究。
总共540例STS患者（男/女：295/245，中位年龄：42岁）。
采用5-T MRI的T1WI序列和脂肪抑制T2加权（T2FS）序列。
手动勾勒感兴趣区域（ROI）以生成影像组学特征。采用自动分割和预训练卷积神经网络（CNN）进行深度学习（DL）分析。去除CNN顶部的最后一个全连接层，并添加全局最大池化将特征图转换为数值。肿瘤分级通过组织学标本确定。
所有基于MRI的模型均采用支持向量机作为分类器。DL特征源自曲线下面积（AUC）最高的DL-MRI模型。将显著的临床变量、肿瘤位置和大小与影像组学和DL特征相结合，构建临床-MRI列线图以识别肿瘤分级。通过Cox比例风险模型评估临床变量和这些基于MRI的特征对总生存期（OS）的预后价值。
临床-MRI鉴别列线图在训练队列中的AUC为0.870，在验证队列中的AUC为0.855，准确率为79.01%，敏感性为79.03%，特异性为78.95%。预后模型在训练和验证队列中对OS显示出良好性能，3年C指数分别为0.681和0.642，5年C指数分别为0.722和0.676。
基于MRI的CAD列线图在低级别和高级别STS分类中表现出有效能力。基于MRI的预后模型在术前具有良好能力，可识别STS的长期生存情况。
3级 技术效能：4级
