Echocardiography-based AI for detection and quantification of atrial septal defect

Frontiers in Cardiovascular Medicine

Cardiovascular Medicine

Haitao Pu¹  Peifang Zhang¹  Yixin Chen¹  Xiaotian Chen¹  Daniel Burkhoff²  Qiushuang Wang³  Liwei Zhang³  Wenxiu Li⁴  Xin Li⁵  Wenjun Wang⁶  Kunlun He⁶  Feifei Yang⁶  Yujiao Deng⁶  Xixiang Lin⁷  Xiao Wang⁷  Xu Chen⁷  Dong Luo⁷ 

[1] BioMind Technology, Beijing, China;Cardiovascular Research Foundation, New York, NY, United States;Department of Cardiology, The Fourth Medical Center of Chinese PLA General Hospital, Beijing, China;Department of Pediatric Cardiac Center, Beijing Anzhen Hospital, Capital Medical University, Beijing, China;Department of Ultrasonography, The Sixth Medical Center of Chinese PLA General Hospital, Beijing, China;Medical Big Data Center, Chinese PLA General Hospital, Beijing, China;Medical Big Data Center, Chinese PLA General Hospital, Beijing, China;Medical School of Chinese PLA, Beijing, China;

关键词: artificial intelligence;    deep learning;    echocardiography;    atrial septal defects;    congenital heart disease;   

DOI  :  10.3389/fcvm.2023.985657

 received in 2022-07-04, accepted in 2023-02-21,  发布年份 2023

来源: Frontiers

【 摘 要 】

ObjectivesWe developed and tested a deep learning (DL) framework applicable to color Doppler echocardiography for automatic detection and quantification of atrial septal defects (ASDs).BackgroundColor Doppler echocardiography is the most commonly used non-invasive imaging tool for detection of ASDs. While prior studies have used DL to detect the presence of ASDs from standard 2D echocardiographic views, no study has yet reported automatic interpretation of color Doppler videos for detection and quantification of ASD.MethodsA total of 821 examinations from two tertiary care hospitals were collected as the training and external testing dataset. We developed DL models to automatically process color Doppler echocardiograms, including view selection, ASD detection and identification of the endpoints of the atrial septum and of the defect to quantify the size of defect and the residual rim.ResultsThe view selection model achieved an average accuracy of 99% in identifying four standard views required for evaluating ASD. In the external testing dataset, the ASD detection model achieved an area under the curve (AUC) of 0.92 with 88% sensitivity and 89% specificity. The final model automatically measured the size of defect and residual rim, with the mean biases of 1.9 mm and 2.2 mm, respectively.ConclusionWe demonstrated the feasibility of using a deep learning model for automated detection and quantification of ASD from color Doppler echocardiography. This model has the potential to improve the accuracy and efficiency of using color Doppler in clinical practice for screening and quantification of ASDs, that are required for clinical decision making.

【 授权许可】

Unknown   
© 2023 Lin, Yang, Chen, Chen, Wang, Li, Wang, Zhang, Li, Deng, Pu, Chen, Wang, Luo, Zhang, Burkhoff and He.