Application of neural networks in prenatal diagnosis of atrioventricular septal defect

Xiaoxue Zhou; Tingyang Yang; Yanping Ruan; Ye Zhang; Xiaowei Liu; Ying Zhao; Xiaoyan Gu; Xinxin Xu; Jiancheng Han; Yihua He

doi:10.21037/tp-23-394

Application of neural networks in prenatal diagnosis of atrioventricular septal defect

Transl Pediatr. 2024 Jan 29;13(1):26-37. doi: 10.21037/tp-23-394. Epub 2024 Jan 15.

Authors

Xiaoxue Zhou^#¹, Tingyang Yang^#², Yanping Ruan¹, Ye Zhang¹, Xiaowei Liu¹, Ying Zhao¹, Xiaoyan Gu¹, Xinxin Xu³, Jiancheng Han¹, Yihua He¹

Affiliations

¹ Maternal-Fetal Consultation Center of Congenital Heart Disease, Department of Echocardiography, Beijing Anzhen Hospital, Capital Medical University, Beijing, China.
² State Key Laboratory of Software Development Environment, Beihang University, Beijing, China.
³ Department of Ultrasound, Hebei Petrochina Central Hospital, Langfang, China.

^# Contributed equally.

Abstract

Background: There is no relevant study on landmarks detection, one of the Convolutional Neural Network algorithms, in the field of fetal echocardiography (FE). This study aimed to explore whether automatic landmarks detection could be used in FE correctly and whether the atrial length (AL) to ventricular length (VL) ratio (AVLR) could be used to diagnose atrioventricular septal defect (AVSD) prenatally.

Methods: This was an observational study. Two hundred and seventy-eight four-chamber views in end diastole, divided into the normal, AVSD, and differential diagnosis groups, were retrospectively included in this study. Seven landmarks were labeled sequentially by the experts on these images, and all images were divided into the training and test sets for normal, AVSD, and differential diagnosis groups. U-net, MA-net, and Link-net were used as landmark prediction neural networks. The accuracy of the landmark detection, AL, and VL measurements, as well as the prenatal diagnostic effectiveness of AVLR for AVSD, was compared with the expert labeled.

Results: U-net, MA-net, and Link-net could detect the landmarks precisely (within the localization error of 0.09 and 0.13 on X and Y axis) and measure AL and VL accurately (the measured pixel distance error of AL and VL were 0.12 and 0.01 separately). AVLR in AVSD was greater than in other groups (P<0.0001), but the statistical difference was not obvious in the complete, partial, and transitional subgroups (P>0.05). The diagnostic effectiveness of AVLR calculated by three models, area under receiver operating characteristic curve could reach 0.992 (0.968-1.000), was consistent with the expert labeled.

Conclusions: U-net, Link-net, and MA-net could detect landmarks and make the measurements accurately. AVLR calculated by three neural networks could be used to make the prenatal diagnosis of AVSD.

Keywords: Fetal echocardiography (FE); atrioventricular septal defect (AVSD); landmark detection; neural networks; prenatal diagnosis.