Segmentation of Five Components in Four Chamber View of Fetal Echocardiography

Yang, Tingyang; Han, Jiancheng; Zhu, Haogang; Li, Tiantian; Liu, Xiaowei; Gu, Xiaoyan; Liu, Xiangyu; An, Shan; Zhang, Ye; He, Yihua

doi:10.1109/isbi45749.2020.9098726

Cited by 15 publications

(18 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, we found U-NET to be the most frequently utilized segmentation model as a baseline of the framework or as optimization. This was seen in (n = 23, 21.50%) ( Aji et al., 2019 ; Arnaout et al., 2021 ; Budd et al., 2019 ; Cerrolaza et al., 2018 ; Chen et al., 2020a ; Deepika et al., 2021 ; Desai et al., 2020 ; Dozen et al., 2020 ; Huang et al., 2018 ; Kim et al., 2018 , 2019a ; Lin et al., 2019a ; Liu et al., 2019 ; Perez-Gonzalez et al., 2020 ; Prieto et al., 2021 ; Qiao and Zulkernine, 2020 ; Singh et al., 2021b ; Weerasinghe et al., 2021 ; Xie et al., 2020a ; Xu et al., 2020a , 2020b ; Yang et al., 2020a ; Yang et al., 2020b ). Other segmentation models were also used, such as DeepLabV3 in ( Yang et al., 2020a ), LinkNet in ( Fathimuthu Joharah and Mohideen, 2020 ), and the Encoder-Decoder network based on VGG16 in ( Li et al., 2017 ).…”

Section: Discussionmentioning

confidence: 86%

“…2D US images were utilized by multi-class segmentation to identify heart disease, including: left heart syndrome (HLHS), total anomalous pulmonary venous connection (TAPVC), pulmonary atresia with intact ventricular septum (PA/IVS), endocardial cushion defect (ECD), fetal cardiac rhabdomyoma (FCR), and Ebstein’s anomaly (EA) ( Yang et al., 2020a ).…”

Section: Resultsmentioning

confidence: 99%

“…Of these studies (n = 13, 12.14%), we did not find any public dataset available online. However, within the private dataset, the image augmentation was applied in (n = 4, 3.7%) studies ( Arnaout et al., 2021 ; Pu et al., 2021 ; Sundaresan et al., 2017 ; Yang et al., 2020a ). The k-fold cross-validation was employed in ( Arnaout et al., 2021 ; Dong et al., 2020 ; Dozen et al., 2020 ; Gong et al., 2020 ; Xu et al., 2020a ).…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Toward deep observation: A systematic survey on artificial intelligence techniques to monitor fetus via ultrasound images

et al. 2022

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 86%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Toward deep observation: A systematic survey on artificial intelligence techniques to monitor fetus via ultrasound images

et al. 2022

View full text Add to dashboard Cite

“…Traditional techniques such as active appearance models 1 and region growing 2 have been used to segment the left and right ventricles. With deep learning approaches, 3,4 more anatomical structures can be segmented. However, due to the large annotation effort required, frequently the clinician demarkings have been restricted to chambers and aorta.…”

Section: Introductionmentioning

confidence: 99%

“…However, due to the large annotation effort required, frequently the clinician demarkings have been restricted to chambers and aorta. 4 For a more complete characterization of the over 18 different types of structural heart defects such as tetralogy of Fallot, atrial septal defects, and ventricular septal defects, 5 a more thorough characterization of the heart regions would be needed. Fortunately, the national guidelines have recommended restricting to specific cardiac views such as the three-vessel trachea view (3VTV) and 4CHV for screening ultrasound examination.…”

Section: Introductionmentioning

confidence: 99%

Multiview and Multiclass Image Segmentation using Deep Learning in Fetal Echocardiography

Wong,

Sinkovskaya,

Abuhamad

et al. 2021

Preprint

View full text Add to dashboard Cite

Congenital heart disease (CHD) is the most common congenital abnormality associated with birth defects in the United States. Despite training efforts and substantial advancement in ultrasound technology over the past years, CHD remains an abnormality that is frequently missed during prenatal ultrasonography. Therefore, computeraided detection of CHD can play a critical role in prenatal care by improving screening and diagnosis. Since many CHDs involve structural abnormalities, automatic segmentation of anatomical structures is an important step in the analysis of fetal echocardiograms. While existing methods mainly focus on the four-chamber view with a small number of structures, here we present a more comprehensive deep learning segmentation framework covering 14 anatomical structures in both three-vessel trachea and four-chamber views. Specifically, our framework enhances the V-Net with spatial dropout, group normalization, and deep supervision to train a segmentation model that can be applied on both views regardless of abnormalities. By identifying the pitfall of using the Dice loss when some labels are unavailable in some images, this framework integrates information from multiple views and is robust to missing structures due to anatomical anomalies, achieving an average Dice score of 79%.

show abstract

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

et al. 2022

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Segmentation of Five Components in Four Chamber View of Fetal Echocardiography

Cited by 15 publications

References 5 publications

Toward deep observation: A systematic survey on artificial intelligence techniques to monitor fetus via ultrasound images

Toward deep observation: A systematic survey on artificial intelligence techniques to monitor fetus via ultrasound images

Multiview and Multiclass Image Segmentation using Deep Learning in Fetal Echocardiography

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

Contact Info

Product

Resources

About