Kamel Peter I, Yi Paul H, Sair Haris I, Lin Cheng Ting
Russell H. Morgan Department of Radiology, The Johns Hopkins University School of Medicine, 601 N Caroline St, Baltimore, MD 21205-2105.
Radiol Cardiothorac Imaging. 2021 Jun 17;3(3):e200486. doi: 10.1148/ryct.2021200486. eCollection 2021 Jun.
To assess the ability of deep convolutional neural networks (DCNNs) to predict coronary artery calcium (CAC) and cardiovascular risk on chest radiographs.
In this retrospective study, 1689 radiographs in patients who underwent cardiac CT and chest radiography within the same year, between 2013 and 2018, were included (mean age, 56 years ± 11 [standard deviation]; 969 radiographs in women). Agatston scores were used as ground truth labels for DCNN training on radiographs. DCNNs were trained for binary classification of nonzero or zero total calcium scores, presence or absence of calcium in each coronary artery, and total calcium scores above or below varying thresholds. Results from classification of test images were compared with established 10-year atherosclerotic cardiovascular disease (ASCVD) risk scores in each cohort. Classifier performance was measured using area under the receiver operating characteristic curve (AUC) with attention maps to highlight areas of decision-making.
Binary classification between zero and nonzero total calcium scores reached an AUC of 0.73 on frontal radiographs, with similar performance on laterals (AUC, 0.70; = .56). Performance was similar for binary classification of absolute total calcium score above or below 100 (AUC, 0.74). Frontal radiographs that tested positive for a predicted nonzero CAC score correlated with a higher 10-year ASCVD risk of 17.2% ± 10.9 compared with 11.9% ± 10.2 for a negative test, indicating predicted CAC score of zero ( < .001). Multivariate logistic regression demonstrated the algorithm could predict a nonzero calcium score independent of traditional cardiovascular risk factors. Performance was reduced for individual coronary arteries. Heat maps primarily localized to the cardiac silhouette and occasionally other cardiovascular findings.
DCNNs trained on chest radiographs had modest accuracy for predicting the presence of CAC correlating with cardiovascular risk. Coronary Arteries, Cardiac, Calcifications/Calculi, Neural NetworksSee also the commentary by Gupta and Blankstein in this issue.©RSNA, 2021.
评估深度卷积神经网络（DCNN）在胸部X线片上预测冠状动脉钙化（CAC）和心血管风险的能力。
在这项回顾性研究中，纳入了2013年至2018年期间同年接受心脏CT和胸部X线检查的患者的1689张X线片（平均年龄56岁±11[标准差]；女性969张X线片）。阿加斯顿评分用作X线片上DCNN训练的真实标签。DCNN针对总钙评分非零或零、各冠状动脉有无钙化以及总钙评分高于或低于不同阈值进行二元分类训练。将测试图像的分类结果与各队列中既定的10年动脉粥样硬化性心血管疾病（ASCVD）风险评分进行比较。使用受试者操作特征曲线下面积（AUC）并结合注意力图来突出决策区域，以衡量分类器性能。
在正位X线片上，总钙评分零与非零之间的二元分类AUC达到0.73，侧位片上表现相似（AUC，0.70；P =.56）。绝对总钙评分高于或低于100的二元分类性能相似（AUC，0.74）。预测非零CAC评分呈阳性的正位X线片与10年ASCVD风险较高相关，为17.2%±10.9，而预测CAC评分为零的阴性测试为11.9%±10.2（P <.001）。多变量逻辑回归表明，该算法可以独立于传统心血管危险因素预测非零钙评分。各冠状动脉的性能有所下降。热图主要定位于心脏轮廓，偶尔也定位于其他心血管表现。
基于胸部X线片训练的DCNN在预测与心血管风险相关的CAC存在方面具有中等准确性。冠状动脉、心脏、钙化/结石、神经网络另见本期Gupta和Blankstein的评论。©RSNA，2021。
