Prenatal diagnosis of tetralogy of Fallot (TOF) and hypoplastic left heart syndrome (HLHS), two serious congenital heart defects, improves outcomes and can in some cases facilitate in utero interventions. In practice, however, the fetal diagnosis rate for these lesions is only 30-50 percent in community settings. Improving fetal diagnosis of congenital heart disease is therefore critical. Deep learning is a cutting-edge machine learning technique for finding patterns in images but has not yet been applied to prenatal diagnosis of congenital heart disease. Using 685 retrospectively collected echocardiograms from fetuses 18-24 weeks of gestational age from 2000-2018, we trained convolutional and fully-convolutional deep learning models in a supervised manner to (i) identify the five canonical screening views of the fetal heart and (ii) segment cardiac structures to calculate fetal cardiac biometrics. We then trained models to distinguish by view between normal hearts, TOF, and HLHS. In a holdout test set of images, F-score for identification of the five most important fetal cardiac views was 0.95. Binary classification of unannotated cardiac views of normal heart vs. TOF reached an overall sensitivity of 75% and a specificity of 76%, while normal vs. HLHS reached a sensitivity of 100% and specificity of 90%, both well above average diagnostic rates for these lesions. Furthermore, segmentation-based measurements for cardiothoracic ratio (CTR), cardiac axis (CA), and ventricular fractional area change (FAC) were compatible with clinically measured metrics for normal, TOF, and HLHS hearts. Thus, using guideline-recommended imaging, deep learning models can significantly improve detection of fetal congenital heart disease compared to the common standard of care.
Trusted Multi-View Deep Learning Classification of Fetal Congenital Heart Disease with Feature-level and Decision-level Fusion
School of Biomedical Engineering, Shanghai Jiao Tong University（生物医学工程学院，上海交通大学）；Shanghai Key Laboratory of Embryo Original Diseases（上海胚胎原发疾病重点实验室）；Shanghai Municipal Key Clinical Specialty（上海市级重点临床专科）
Iterative Segmentation from Limited Training Data: Applications to Congenital Heart Disease
Martinos Center for Biomedical Imaging, Massachusetts General Hospital（马萨诸塞州总医院马丁诺斯生物医学成像中心）；Harvard Medical School（哈佛医学院）；Cornell University（康奈尔大学）；Philips Research Laboratories（飞利浦研究实验室）；Boston Children’s Hospital（波士顿儿童医院）
TPA: Temporal Prompt Alignment for Fetal Congenital Heart Defect Classification
Department of Machine Learning Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) Abu Dhabi, UAE（机器学习系，Mohamed bin Zayed人工智能大学（MBZUAI），阿布扎比，阿联酋）；Department of Computer Vision Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) Abu Dhabi, UAE（计算机视觉系，Mohamed bin Zayed人工智能大学（MBZUAI），阿布扎比，阿联酋）
Real-time Cardiovascular MR with Spatio-temporal Artifact Suppression using Deep Learning - Proof of Concept in Congenital Heart Disease
University College London（伦敦大学学院）；Centrum Wiskunde & Informatica (CWI)（数学和计算机科学中心（CWI））；UCL Centre for Cardiovascular Imaging（伦敦大学学院心血管影像中心）；Institute of Cardiovascular Science（心血管科学研究所）
SDF4CHD: Generative Modeling of Cardiac Anatomies with Congenital Heart Defects
Stanford University（斯坦福大学）；Cardiovascular Institute（心血管研究所）；ETH Zurich and University Zurich（苏黎世联邦理工学院和苏黎世大学）；Institute for Biomedical Engineering（生物医学工程研究所）
Self-supervised Normality Learning and Divergence Vector-guided Model Merging for Zero-shot Congenital Heart Disease Detection in Fetal Ultrasound Videos
University of Oxford（牛津大学）；University of Technology Nuremberg（纽伦堡工业大学）
Automated detection of pediatric congenital heart disease from phonocardiograms using deep and handcrafted feature fusion
Electrical and Computer System Engineering, Monash University（莫纳什大学电气与计算机系统工程系）；Biomedical Engineering, McGill University（麦吉尔大学生物医学工程系）；Electrical and Electronic Engineering (EEE), United International University（国际大学电气与电子工程系）；Healthcare Engineering Innovation Group (HEIG), Department of Biomedical Engineering and Biotechnology, Khalifa University（卡利法大学健康工程创新组，生物医学工程与生物技术系）；Department of Electrical and Computer Engineering (ECE), Duke University（杜克大学电气与计算机工程系）
How well do U-Net-based segmentation trained on adult cardiac magnetic resonance imaging data generalise to rare congenital heart diseases for surgical planning?
Mannheim University of Applied Science（曼海姆应用科技大学）；UT Southwestern Medical Center（德克萨斯大学西南医学中心）；DZHK (German Centre for Cardiovascular Research)（德国心血管研究中心）；German Heart Centre Munich（慕尼黑德国心脏中心）；Hannover Medical School（汉诺威医学院）；Heidelberg University Hospital（海德堡大学医院）
Automated interpretation of congenital heart disease from multi-view echocardiograms
Capital Medical University（首都医科大学）；Carnegie Mellon University（卡内基梅隆大学）；Harvard University（哈佛大学）；Harvard Medical School（哈佛医学院）；Beijing Children’s Hospital（北京儿童医院）；National Center for Children’s Health（国家儿童健康中心）；Harbin Engineering University（哈尔滨工程大学）；National Research Institute for Family Planning（国家计划生育委员会科学技术研究所）
Orientation-Robust Latent Motion Trajectory Learning for Annotation-free Cardiac Phase Detection in Fetal Echocardiography
Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford（生物医学工程研究所，工程科学系，牛津大学）；Nuffield Department of Women’s and Reproductive Health, University of Oxford（妇女与生殖健康纽菲尔德部门，牛津大学）
EllipseNet: Anchor-Free Ellipse Detection for Automatic Cardiac Biometrics in Fetal Echocardiography
Peng Cheng Laboratory（鹏城实验室）；Beihang University（北京航空航天大学）；Capital Medical University（首都医科大学）；Beijing Anzhen Hospital（北京安贞医院）；Beijing Advanced Innovation Center for Big Data-Based Precision Medicine（北京大数据精准医疗先进创新中心）
DCD: A Semantic Segmentation Model for Fetal Ultrasound Four-Chamber View
School of Information and Communication（信息与通信学院）；Guilin University of Electronic Technology（桂林电子科技大学）；Faculty of Humanities and Arts（人文艺术学院）；Macau University of Science and Technology（澳门科技大学）；Guangxi Key Laboratory of Machine Vision and Intelligent Control（广西机器视觉与智能控制重点实验室）；Wuzhou University（梧州大学）；Faculty of Innovation Engineering（创新工程学院）；Department of Public Education（公共教育系）；Wuzhou Medical College（梧州医学院）；College of Big Data and Software Engineering（大数据与软件工程学院）；Shenzhen Maternity and Child Healthcare Hospital（深圳妇幼保健医院）；Southern Medical University（南方医科大学）；School of Microelectronics（微电子学院）
CardiacNet: Learning to Reconstruct Abnormalities for Cardiac Disease Assessment from Echocardiogram Videos
The Hong Kong University of Science and Technology（香港科技大学）；Guangdong Provincial People’s Hospital（广东省人民医院）；HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute（香港科技大学深港协同创新研究院）
Atrial Septal Defect Detection in Children Based on Ultrasound Video Using Multiple Instances Learning
Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiaotong University（上海交通大学医学院附属上海儿童医学中心）；Minhang District Centers for Disease Control and Prevention（闵行区疾病预防控制中心）；Xi’an Jiaotong-Liverpool University（西交利物浦大学）；East China Normal University（华东师范大学）；University of Shanghai for Science and Technology（上海理工大学）
Deformable Image Registration for Self-supervised Cardiac Phase Detection in Multi-View Multi-Disease Cardiac Magnetic Resonance Images
Department of Internal Medicine III, Heidelberg University Hospital（海德堡大学第三内科部门医院）；Medical Faculty of Heidelberg University（海德堡大学医学院）；DZHK (German Centre for Cardiovascular Research)（德国心血管研究中心）；Division of Pediatric Cardiology, Department of Pediatrics, UT Southwestern /Children’s Health（儿童心脏病科，德克萨斯西南医学中心/儿童健康）；Department of Cardiothoracic, Transplantation and Vascular Surgery, Hannover Medical School（汉诺威医学院心胸外科、移植外科和血管外科部门）
本文提出一种自监督深度学习方法，通过多视角多疾病心脏磁共振图像检测五个关键帧，提高了心脏相位检测的准确率，适用于SAX和4CH cine CMR。
