Diagnosis of fetal total anomalous pulmonary venous connection based on the post‐left atrium space ratio using artificial intelligence

• Published in: Prenatal diagnosis Vol. 42; no. 10; pp. 1323 - 1331
• Main Authors: Wang, Xin, Yang, Ting‐Yang, Zhang, Ying‐Ying, Liu, Xiao‐Wei, Zhang, Ye, Sun, Lin, Gu, Xiao‐Yan, Chen, Zhuo, Guo, Yong, Xue, Chao, Han, Jian‐Cheng, Zhu, Hao‐Gang, He, Yi‐Hua
• Format: Journal Article
• Language: English
• Published: Charlottesville Wiley Subscription Services, Inc 01.09.2022
• Subjects: Annotations; Aorta; Apexes; Artificial intelligence; Atria; Diagnosis; Echocardiography; Epicardium; Fetuses; Gestational age; Medical diagnosis; Prenatal diagnosis; Segmentation; Test sets
• ISSN: 0197-3851; 1097-0223; 1097-0223
• DOI: 10.1002/pd.6220

Objective To explore whether the post‐left atrium space (PLAS) ratio would be useful for prenatal diagnosis of total anomalous pulmonary venous connection (TAPVC) using echocardiography and artificial intelligence. Methods We retrospectively included 642 frames of four‐chamber views from 319 fetuses (32 with TAPVC and 287 without TAPVC) in end‐systolic and end‐diastolic periods with multiple apex directions. The average gestational age was 25.6 ± 2.7 weeks. No other cardiac or extracardiac malformations were observed. The dataset was divided into a training set (n = 540; 48 with TAPVC and 492 without TAPVC) and test set (n = 102; 20 with TAPVC and 82 without TAPVC). The PLAS ratio was defined as the ratio of the epicardium‐descending aortic distance to the center of the heart‐descending aortic distance. Supervised learning was used in DeepLabv3+, FastFCN, PSPNet, and DenseASPP segmentation models. The area under the curve (AUC) was used on the test set. Results Expert annotations showed that this ratio was not related to the period or apex direction. It was higher in the TAPVC group than in the control group detected by the expert and the four models. The AUC of expert annotations, DeepLabv3+, FastFCN, PSPNet, and DenseASPP were 0.977, 0.941, 0.925, 0.856, and 0.887, respectively. Conclusion Segmentation models achieve good diagnostic accuracy for TAPVC based on the PLAS ratio. Key points What's already known about this topic? Obstructed total anomalous pulmonary venous connection (TAPVC) is a life‐threatening neonatal emergency requiring immediate surgery after birth. But the detection rate of TAPVC remains suboptimal. Convolutional neural network (CNN) can be useful for the acquisition of standard cardiac image planes, automated cardiac measurements, assessment of the quality of standard views, and classification of fetal hearts into normal or abnormal hearts. However, there was no CNN used to diagnose fetal TAPVC. What does this study add? The post‐left atrium space (PLAS) ratio is a sensitive parameter for the prenatal diagnosis of TAPVC. Semantic segmentation models achieved a diagnostic accuracy comparable to that of clinical experts for the diagnosis of TAPVC based on the PLAS ratio.