Convolutional Neural Network Using a Breast MRI Tumor Dataset Can Predict Oncotype Dx Recurrence Score.
Convolutional Neural Network Using a Breast MRI Tumor Dataset Can Predict Oncotype Dx Recurrence Score.
复制标题
DOI：
10.1002/jmri.26244
复制
发表时间：
2019-03
期刊：
影响因子：
--
通讯作者：
Jambawalikar S
中科院分区：
文献类型：
--
作者：
Ha R;Chang P;Mutasa S;Karcich J;Goodman S;Blum E;Kalinsky K;Liu MZ;Jambawalikar S
Oncotype Dx是一种经过验证的遗传分析，可提供复发评分（RS），以定量预测符合雌激素受体阳性/人表皮生长因子受体-2阴性(ER+/HER2−)/淋巴结阴性侵袭性乳腺癌患者的预后。虽然有效，但该检测是侵入性的且昂贵的，这促使这项研究确定放射组学的潜在作用。我们假设卷积神经网络（CNN）可以使用MRI数据集预测Oncotype Dx RS。机构审查委员会（IRB）于2010年1月至2016年6月批准的回顾性研究。总共有134例ER+/HER2−浸润性导管癌患者接受了乳房MRI和Oncotype Dx RS评估。将患者分为低危组（1组，RS <18）、中危组（2组，RS 18 - 30）、高危组（3组，RS 18 - 30）。1.5T和3.0T。乳腺MRI，对比后T1。对每个乳腺肿瘤进行三维分割。总共评估了134个肿瘤的1649块体积切片（平均12.3片/个肿瘤）。一个CNN由四个卷积层和最大池化层组成。50%的Dropout应用于倒数第二个完全连接的层，以防止过拟合。采用三级预测（1组vs. 2组vs. 3组）和两级预测（1组vs. 2/3组）模型。采用80%训练和20%测试进行5倍交叉验证检验。评估诊断的准确性、敏感性、特异性和受试者工作特征（ROC）曲线下面积（AUC）。CNN在三类预测中总体准确率为81%(95%置信区间[CI]±4%)，特异性为90% (95% CI±5%)，敏感性为60% (95% CI±6%)，ROC曲线下面积为0.92 （SD, 0.01）。CNN在两类预测中总体准确率为84% (95% CI±5%)，特异性为81% (95% CI±4%)，敏感性为87% (95% CI±5%)，ROC曲线下面积为0.92 （SD, 0.01）。对目前的深度CNN架构进行训练来预测Oncotype DX RS是可行的。
Oncotype Dx is a validated genetic analysis that provides a recurrence score (RS) to quantitatively predict outcomes in patients who meet the criteria of estrogen receptor positive / human epidermal growth factor receptor-2 negative (ER+/HER2−)/node negative invasive breast carcinoma. Although effective, the test is invasive and expensive, which has motivated this investigation to determine the potential role of radiomics. We hypothesized that convolutional neural network (CNN) can be used to predict Oncotype Dx RS using an MRI dataset. Institutional Review Board (IRB)-approved retrospective study from January 2010 to June 2016. In all, 134 patients with ER+/HER2− invasive ductal carcinoma who underwent both breast MRI and Oncotype Dx RS evaluation. Patients were classified into three groups: low risk (group 1, RS <18), intermediate risk (group 2, RS 18–30), and high risk (group 3, RS >30). 1.5T and 3.0T. Breast MRI, T1 postcontrast. Each breast tumor underwent 3D segmentation. In all, 1649 volumetric slices in 134 tumors (mean 12.3 slices/tumor) were evaluated. A CNN consisted of four convolutional layers and max-pooling layers. Dropout at 50% was applied to the second to last fully connected layer to prevent overfitting. Three-class prediction (group 1 vs. group 2 vs. group 3) and two-class prediction (group 1 vs. group 2/3) models were performed. A 5-fold crossvalidation test was performed using 80% training and 20% testing. Diagnostic accuracy, sensitivity, specificity, and receiver operating characteristic (ROC) area under the curve (AUC) were evaluated. The CNN achieved an overall accuracy of 81% (95% confidence interval [CI] ± 4%) in three-class prediction with specificity 90% (95% CI ± 5%), sensitivity 60% (95% CI ± 6%), and the area under the ROC curve was 0.92 (SD, 0.01). The CNN achieved an overall accuracy of 84% (95% CI ± 5%) in two-class prediction with specificity 81% (95% CI ± 4%), sensitivity 87% (95% CI ± 5%), and the area under the ROC curve was 0.92 (SD, 0.01). It is feasible for current deep CNN architecture to be trained to predict Oncotype DX RS.
登录
查看更多内容
在美国临床实践中，对乳腺癌的21基因复发评分测定法的利用和影响：从2010年到2012年的国家癌症数据库分析中学到的经验教训。
影响因子：
3.8
作者：
Orucevic A;Heidel RE;Bell JL
通讯作者：
Bell JL
影响因子：
3.9
作者：
Mandic, DP
通讯作者：
Mandic, DP
影响因子：
19.7
作者：
Li H;Zhu Y;Burnside ES;Drukker K;Hoadley KA;Fan C;Conzen SD;Whitman GJ;Sutton EJ;Net JM;Ganott M;Huang E;Morris EA;Perou CM;Ji Y;Giger ML
通讯作者：
Giger ML
影响因子：
4
作者：
Song JL;Chen C;Yuan JP;Sun SR
通讯作者：
Sun SR
影响因子：
19.7
作者：
Ashraf AB;Daye D;Gavenonis S;Mies C;Feldman M;Rosen M;Kontos D
通讯作者：
Kontos D