An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease.
An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease.
复制标题
DOI：
10.1038/s41591-021-01342-5
复制
发表时间：
2021-05
期刊：
影响因子：
82.9
通讯作者：
Moon-Grady AJ
中科院分区：
文献类型：
--
作者：
Arnaout R;Curran L;Zhao Y;Levine JC;Chinn E;Moon-Grady AJ
先天性心脏病（CHD）是最常见的出生缺陷。胎儿筛查超声可提供心脏的五个视图，总共可以检测出 90% 的复杂 CHD，但实际上，灵敏度低至 30%。在这里，我们使用来自 1,326 幅回顾性超声心动图和 18 至 24 周胎儿超声筛查的 107,823 幅图像，训练了一组神经网络来识别推荐的心脏视图并区分正常心脏和复杂的先心病。我们还使用分割模型来计算标准胎儿心胸测量值。在 4,108 项胎儿调查（0.9% CHD，> 440 万张图像）的内部测试集中，该模型在区分正常心脏和异常心脏方面实现了 0.99 的曲线下面积 (AUC)、95% 的敏感性（95% 置信区间 (CI)，84-99%）、96% 的特异性（95% CI，95-97%）和 100% 的阴性预测值。模型的敏感性与临床医生相当，并且在医院外和较低质量的图像上保持稳健。该模型的决策基于临床相关特征。心脏测量值与报告的正常和异常心脏测量值相关。应用于指南推荐的成像，集成学习模型可以显着改善胎儿冠心病的检测，这是一项关键的全球诊断挑战。
Congenital heart disease (CHD) is the most common birth defect. Fetal screening ultrasound provides five views of the heart that together can detect 90% of complex CHD, but in practice, sensitivity is as low as 30%. Here, using 107,823 images from 1,326 retrospective echocardiograms and screening ultrasounds from 18- to 24-week fetuses, we trained an ensemble of neural networks to identify recommended cardiac views and distinguish between normal hearts and complex CHD. We also used segmentation models to calculate standard fetal cardiothoracic measurements. In an internal test set of 4,108 fetal surveys (0.9% CHD, >4.4 million images), the model achieved an area under the curve (AUC) of 0.99, 95% sensitivity (95% confidence interval (CI), 84–99%), 96% specificity (95% CI, 95–97%) and 100% negative predictive value in distinguishing normal from abnormal hearts. Model sensitivity was comparable to that of clinicians and remained robust on outside-hospital and lower-quality images. The model’s decisions were based on clinically relevant features. Cardiac measurements correlated with reported measures for normal and abnormal hearts. Applied to guideline-recommended imaging, ensemble learning models could significantly improve detection of fetal CHD, a critical and global diagnostic challenge.
登录
查看更多内容
对先天性心脏缺陷的婴儿的产前诊断与存活率关联的基于人群的研究。
影响因子：
2.8
作者：
Oster, Matthew E.;Kim, Christopher H.;Kusano, Aaron S.;Cragan, Janet D.;Dressler, Paul;Hales, Alice R.;Mahle, William T.;Correa, Adolfo
通讯作者：
Correa, Adolfo
影响因子：
4.3
作者：
Chu, Chen;Yan, Yingliu;Gui, Yonghao
通讯作者：
Gui, Yonghao
DOI：
10.1002/j.2205-0140.2015.tb00024.x
发表时间：
2015-02-01
影响因子：
--
作者：
Miceli, Frances
通讯作者：
Miceli, Frances
影响因子：
168.9
作者：
Attia, Zachi, I;Noseworthy, Peter A.;Friedman, Paul A.
通讯作者：
Friedman, Paul A.
影响因子：
64.8
作者：
Ouyang D;He B;Ghorbani A;Yuan N;Ebinger J;Langlotz CP;Heidenreich PA;Harrington RA;Liang DH;Ashley EA;Zou JY
通讯作者：
Zou JY