The DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Expert-level prenatal detection of complex congenital heart disease from screening ultrasound using deep learning
Open accessAbstract
Congenital heart disease (CHD) is the most common birth defect. Fetal survey ultrasound is recommended worldwide, including five views of the heart that together could detect 90% of complex CHD. In practice, however, sensitivity is as low as 30%. We hypothesized poor detection results from challenges in acquiring and interpreting diagnostic-quality cardiac views, and that deep learning could improve complex CHD detection. Using 107,823 images from 1,326 retrospective echocardiograms and surveys from 18-24 week fetuses, we trained an ensemble of neural networks to (i) identify recommended cardiac views and (ii) distinguish between normal hearts and complex CHD. Finally, (iii) we used segmentation models to calculate standard fetal cardiothoracic measurements. In a test set of 4,108 fetal surveys (0.9% CHD, >4.4 million images, about 400 times the size of the training dataset) the model achieved an AUC of 0.99, 95% sensitivity (95%CI, 84-99), 96% specificity (95%CI, 95-97), and 100% NPV in distinguishing normal from abnormal hearts. Sensitivity was comparable to clinicians’ task-for-task and remained robust on external and lower-quality images. The model’s decisions were based on clinically relevant features. Cardiac measurements correlated with reported measures for normal and abnormal hearts. Applied to guidelines-recommended imaging, ensemble learning models could significantly improve detection of fetal CHD and expand telehealth options for prenatal care at a time when the COVID-19 pandemic has further limited patient access to trained providers. This is the first use of deep learning to ∼double standard clinical performance on a critical and global diagnostic challenge.
Verification finished with gaps
Some checks didn’t finish. The results below cover the steps that did.
- Registry recordVerified
- Full textRead the full text
- Reference list51 references found
- Reference check53 of 53 references checked
- StatisticsRecomputed
- ClaimsDidn’t finish — the rest of the report still applies.
- Cited sourcesDidn’t finish — the rest of the report still applies.
- Self-checkChecked
- VerificationFindings verified
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
An ensemble of deep learning models was trained on fetal echocardiograms and screening ultrasounds to identify recommended cardiac views, distinguish normal hearts from complex congenital heart disease (CHD), and derive cardiothoracic measurements. In a large internal screening-ultrasound test set with population-like CHD prevalence, the model achieved an AUC of 0.99 with 95% sensitivity and 96% specificity.
Key findings
- Using up to 107,823 images from up to 1,326 studies, the ensemble achieved an AUC of 0.99 in OB-4000, a test set of 4,108 fetal surveys with 0.9% CHD and >4.4 million images; sensitivity was 95% (95% CI, 84–99%), specificity 96% (95% CI, 95–97%), positive predictive value 20% (95% CI, 17–23%), and negative predictive value 100%.
- View classification on FETAL-125 had an F score of 0.93 (AUC range, 0.94–0.98); on OB-125, sensitivity was 90% and specificity 78%, rising to 96% and 92% when using images with prediction probabilities at or above the first quartile.
- On external BCH-400 fetal echocardiograms with 92% CHD, composite classification AUC was 0.89; on TWINS-10, sensitivity was 100% and specificity 72%; for specific lesions, sensitivity was 71% for tetralogy of Fallot and 89% for hypoplastic left heart syndrome.
- Segmentation-derived measurements included cardiothoracic ratio, cardiac axis, and fractional area change; for example, mean cardiac axis was increased in tetralogy of Fallot at 63° ± 16° (range, 54–80°; P = 0.007) versus a normal reference of 45° ± 20°.
Methods
- Retrospective, de-identified fetal echocardiograms and second-trimester screening ultrasounds from 18- to 24-week fetuses, performed between 2000 and 2019 at UCSF and Boston Children’s Hospital; fetuses with clinically significant non-cardiac malformations were excluded, and gold-standard normal/CHD labels came from expert review plus electronic health-record codes for studies from 2012 onward.
- The ensemble used a ResNet-based view classifier for five recommended views (3VT, 3VV, LVOT, A4C, ABDO) plus non-target images, per-view binary diagnostic classifiers, a rule-based composite diagnostic classifier, and modified U-Net segmentation models for cardiothoracic structures.
- Evaluation used five independent test sets—FETAL-125, OB-125, OB-4000, BCH-400, and TWINS-10—with metrics including AUC, sensitivity, specificity, predictive values, F score, and Jaccard similarity; model performance was also compared with seven clinicians on a five-image-per-heart task in OB-125.
Limitations
- The paper states that testing was retrospective, that OB-4000 came from a center with fetal screening experience, and that prospective testing on consecutive cases across multiple centers—including rural, community, and non-expert centers—is needed, along with more multiple gestations and imaging with non-cardiac malformations.
- The paper also states that more training data are needed for specific lesion types and segmentation, and that further clinical validation of segmentation-derived fetal biometrics is required, especially for CHD lesions without established metrics.
- Evident from the text: the external BCH-400 test set had a very high CHD prevalence (92%) and lower composite AUC (0.89); the clinician comparison was limited to the 38 of 125 OB-125 surveys containing all five views; and data are not publicly available, with model weights available only upon request.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
This check didn’t finish.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Arnaout et al. used trained neural networks to detect the various cardiac views and distinguish between normal and abnormal images on fetal echocardiogram.77 The authors found an AUC of 0.99, a sensitivity of 95%, and specificity of 96%.”
From Artificial intelligence in obstetric ultrasound: A scoping review. (2023) · cites this paper as [77]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The combination of standard views allows for the detection of up to 90% of CHD; however, in clinical practice, the detection rate is only about 30% [48,50].”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [50]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“summarized that AI applications rather assist than replace human observers [50].”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [50]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“The retrospective study of Huo et al. showed that AI-assisted US improved the accuracy of uterine fibroid assessment of young sonographers”
From Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology (2023) · cites this paper as [50]
Experimental — accuracy not yet measuredIn the citing paper
“The results obtained following previous research using the DL method from fetal US images can help increase the detection of fetal heart abnormalities compared to human experts [10,17,18,24].”
From Deep Learning for Improving the Effectiveness of Routine Prenatal Screening for Major Congenital Heart Diseases (2022) · cites this paper as [24]
In the citing paper
“Worldwide, expert fetal cardiologists can detect as few as 30% to 50% of these CHD conditions before birth [24].”
From Deep Learning for Improving the Effectiveness of Routine Prenatal Screening for Major Congenital Heart Diseases (2022) · cites this paper as [24]
In the citing paper
“However, combining human-performed US and machine analysis allowed researchers to detect 95% CHDs in their test dataset.”
From Deep Learning for Improving the Effectiveness of Routine Prenatal Screening for Major Congenital Heart Diseases (2022) · cites this paper as [24]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Different lower confidence limits for OB-4000 sensitivity
MinorThe abstract and Results report the same 95% sensitivity and 96% specificity for the 4,108-survey test set, but give different lower limits for the sensitivity confidence interval: 84% and 83%.
“In an internal test set of 4,108 fetal surveys (0.9% CHD, >4.4 million images), the model achieved an area under the curve (AUC) of 0.99, 95% sensitivity (95% confidence interval (CI), 84–99%), 96% specificity (95% CI, 95–97%)”
“This allowed a sensitivity of 95% (95% CI, 83–99%), a specificity of 96% (95% CI, 95–97%), a positive predictive value of 20% (95% CI, 17–23%) and a negative predictive value of 100% in OB-4000.”
Citations
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Identity mappings in deep residual networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_38 · Cited by 10,741
Identity Mappings in Deep Residual Networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · arXiv · doi:10.48550/arxiv.1603.05027 · Cited by 10,665
Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., et al. (2016) · JAMA · doi:10.1001/jama.2016.17216 · Cited by 7,697
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01228-7 · Cited by 6,909
Striving for Simplicity: The All Convolutional Net
Springenberg, J. T., Dosovitskiy, A., Brox, T., & Riedmiller, M. (2014) · arXiv · doi:10.48550/arxiv.1412.6806 · Cited by 3,447
Zech, J. R., Badgeley, M. A., Liu, M., Costa, A. B., Titano, J. J., & Oermann, E. K. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002683 · Cited by 1,719 · cited 1×
“To validate that the model generalizes beyond the medical center where it was trained 34 , we tested it on fetal echocardiograms from an unaffiliated, geographically remote medical center with a high prevalence of CHD (BCH-400; Table 1).”
MethodAttia, Z. I., Noseworthy, P. A., Lopez-Jimenez, F., Asirvatham, S. J., Deshmukh, A. J., Gersh, B. J., Carter, R. E., Yao, X., Rabinstein, A. A., Erickson, B. J., et al. (2019) · The Lancet · doi:10.1016/s0140-6736(19)31721-0 · Cited by 1,613
Donofrio, M. T., Moon-Grady, A. J., Hornberger, L. K., Copel, J. A., Sklansky, M. S., Abuhamad, A., Cuneo, B. F., Huhta, J. C., Jonas, R. A., Krishnan, A., et al. (2014) · Circulation · doi:10.1161/01.cir.0000437597.44550.5d · Cited by 1,275
Self-citationVideo-based AI for beat-to-beat assessment of cardiac function
Ouyang, D., He, B., Ghorbani, A., Yuan, N., Ebinger, J., Langlotz, C. P., Heidenreich, P. A., Harrington, R. A., Liang, D. H., Ashley, E. A., et al. (2020) · Nature · doi:10.1038/s41586-020-2145-8 · Cited by 1,153 · cited 1×
“DL has been applied to adult cardiac ultrasound 25,26 , besting clinicians on view classification on small, downsampled datasets 27 .”
MethodU-Net: Convolutional Networks for Biomedical Image SegmentationPreprint
Ronneberger, O., Fischer, P., & Brox, T. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.04597 · Cited by 1,079
Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study
Chilamkurthy, S., Ghosh, R., Tanamala, S., Biviji, M., Campeau, N. G., Venugopal, V. K., Mahajan, V., Rao, P., & Warier, P. (2018) · The Lancet · doi:10.1016/s0140-6736(18)31645-3 · Cited by 1,020
ISUOG Practice Guidelines (updated): sonographic screening examination of the fetal heart
Carvalho, J., Allan, L., Chaoui, R., Copel, J., DeVore, G., Hecher, K., Lee, W., Munoz, H., Paladini, D., Tutschek, B., et al. (2013) · Ultrasound in Obstetrics & Gynecology · doi:10.1002/uog.12403 · Cited by 681
Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist.
Norgeot, B., Quer, G., Beaulieu-Jones, B. K., Torkamani, A., Dias, R., Gianfrancesco, M., Arnaout, R., Kohane, I. S., Saria, S., Topol, E., et al. (2020) · Nature Medicine · doi:10.1038/s41591-020-1041-y · Cited by 665 · cited 1×
“The methods below describe datasets, data processing, algorithms and performance measures 46 .”
MethodSelf-citationFast and accurate view classification of echocardiograms using deep learning
Madani, A., Arnaout, R., Mofrad, M., & Arnaout, R. (2018) · npj Digital Medicine · doi:10.1038/s41746-017-0013-1 · Cited by 585
Self-citationWHO recommendations on antenatal care for a positive pregnancy experience—going beyond survival
Tunçalp, Ö., Peña‐Rosas, J. P., Lawrie, T. A., Bucagu, M., Oladapo, O., Portela, A. G., & Gülmezoglu, A. M. (2017) · BJOG An International Journal of Obstetrics & Gynaecology · doi:10.1111/1471-0528.14599 · Cited by 502
SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound
Baumgartner, C. F., Kamnitsas, K., Matthew, J., Fletcher, T. P., Smith, S., Koch, L. M., Kainz, B., & Rueckert, D. (2017) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2017.2712367 · Cited by 416
Yagel, S., Cohen, S. M., & Achiron, R. (2001) · Ultrasound in Obstetrics and Gynecology · doi:10.1046/j.1469-0705.2001.00414.x · Cited by 299
Prenatal detection of congenital heart disease
Friedberg, M. K., Silverman, N. H., Moon-Grady, A. J., Tong, E., Nourse, J., Sorenson, B., Lee, J., & Hornberger, L. K. (2009) · The Journal of Pediatrics · doi:10.1016/j.jpeds.2009.01.050 · Cited by 299
Self-citationHolland, B. J., Myers, J. A., & Woods, C. R. (2015) · Ultrasound in Obstetrics & Gynecology · doi:10.1002/uog.14882 · Cited by 286
Practice Bulletin No. 175: Ultrasound in Pregnancy
(2016) · Obstetrics & Gynecology · doi:10.1097/aog.0000000000001815 · Cited by 282 · cited 2×
“In practice, however, detection is often as low as 30% 1,13,14 , even where ultrasound is universal 9,10,14 .”
BackgroundPrenatal care redesign: creating flexible maternity care models through virtual care
Peahl, A. F., Smith, R. D., & Moniz, M. H. (2020) · American Journal of Obstetrics and Gynecology · doi:10.1016/j.ajog.2020.05.029 · Cited by 220
Fetal Aortic Valvuloplasty for Evolving Hypoplastic Left Heart Syndrome
Freud, L. R., McElhinney, D. B., Marshall, A. C., Marx, G. R., Friedman, K. G., del Nido, P. J., Emani, S. M., Lafranchi, T., Silva, V., Wilkins-Haug, L. E., et al. (2014) · Circulation · doi:10.1161/circulationaha.114.009032 · Cited by 216
Best, K., & Rankin, J. M. (2016) · Journal of the American Heart Association · doi:10.1161/jaha.115.002846 · Cited by 159
AIUM Practice Parameter for the Performance of Fetal Echocardiography
(2019) · Journal of Ultrasound in Medicine · doi:10.1002/jum.15188 · Cited by 134
AIUM Practice Guideline for the Performance of Fetal Echocardiography
(2011) · Journal of Ultrasound in Medicine · doi:10.7863/jum.2011.30.1.127 · Cited by 124
AIUM Practice Guideline for the Performance of Fetal Echocardiography
(2013) · Journal of Ultrasound in Medicine · doi:10.7863/ultra.32.6.1067 · Cited by 124
Donofrio, M., Moon-Grady, A., Hornberger, L., Copel, J. A., Sklansky, M., Abuhamad, A., Cuneo, B. F., Huhta, J. C., Ra, J., Krishnan, A. N., et al. (2014) · eScholarship (California Digital Library) · Cited by 122
Self-citationOster, M. E., Kim, C. H., Kusano, A. S., Cragan, J. D., Dressler, P. B., Hales, A. R., Mahle, W. T., & Correa, A. (2014) · The American Journal of Cardiology · doi:10.1016/j.amjcard.2013.11.066 · Cited by 91
Sun, H. Y., Proudfoot, J. A., & McCandless, R. T. (2018) · Congenital Heart Disease · doi:10.1111/chd.12648 · Cited by 73
Li, Y.-F., Zhou, K.-Y., Fang, J., Wang, C., Hua, Y.-M., & Mu, D.-Z. (2016) · World Journal of Pediatrics · doi:10.1007/s12519-016-0016-z · Cited by 67
Liu, H., Zhou, J., Feng, Q., Gu, H., Wan, G., Zhang, H.-M., Xie, Y.-J., & Li, X.-S. (2014) · European Journal of Preventive Cardiology · doi:10.1177/2047487314551547 · Cited by 62
Chen, C., Yan, Y., Ren, Y., Li, X., & Gui, Y. (2016) · Acta Obstetricia Et Gynecologica Scandinavica · doi:10.1111/aogs.13085 · Cited by 50
Relation of Prenatal Diagnosis With One-Year Survival Rate for Infants With Congenital Heart Disease
Wright, L., Ehrlich, A. C., Stauffer, N., Samai, C., Kogon, B. E., & Oster, M. E. (2013) · The American Journal of Cardiology · doi:10.1016/j.amjcard.2013.11.065 · Cited by 44
AIUM Practice Parameter for the Performance of a Peripheral Venous Ultrasound Examination
(2020) · Journal of Ultrasound in Medicine · doi:10.1002/jum.15263 · Cited by 40
Lee, S. J., Jacobson, J. A., Kim, S.-M., Fessell, D. P., Jiang, Y., Girish, G., & Magerkurth, O. (2013) · Journal of Ultrasound in Medicine · doi:10.7863/ultra.32.12.2075 · Cited by 37
Prenatal detection of major congenital heart disease – optimising resources to improve outcomes
Corcoran, S. M., Briggs, K., Connor, H. O., Müllers, S., Monteith, C., Donnelly, J. C., Dicker, P., Franklin, O., Malone, F. D., & Breathnach, F. (2016) · European Journal of Obstetrics & Gynecology and Reproductive Biology · doi:10.1016/j.ejogrb.2016.06.008 · Cited by 35
Detection of fetal cardiac anomalies: cost‐effectiveness of increased number of cardiac views
Bak, G. S., Shaffer, B. L., Madriago, E. J., Allen, A. J., Kelly, B., Caughey, A. B., & Pereira, L. R. (2020) · Ultrasound in Obstetrics and Gynecology · doi:10.1002/uog.21977 · Cited by 35
Sklansky, M., & DeVore, G. R. (2016) · Journal of Ultrasound in Medicine · doi:10.7863/ultra.15.07021 · Cited by 31
Neonatal management and outcomes of prenatally diagnosed CHDs
Bensemlali, M., Bajolle, F., Laux, D., Parisot, P., Ladouceur, M., Fermont, L., Lévy, M., Le Bidois, J., Raimondi, F., Ville, Y., et al. (2017) · Cardiology in the Young · doi:10.1017/s1047951116000639 · Cited by 31
Letourneau, K. M., Horne, D., Soni, R. N., McDonald, K. R., Karlicki, F. C., & Fransoo, R. (2017) · Journal of Ultrasound in Medicine · doi:10.1002/jum.14453 · Cited by 26
Accuracy of Prenatal Diagnosis of Congenital Cardiac Malformations
Pinheiro, D. O., Varisco, B. B., da Silva, M. B., Duarte, R. S., Deliberali, G. D., Maia, C. R., Jiménez, M. F., & Beitune, P. E. (2018) · Revista Brasileira Ginecologia e Obstetrícia · doi:10.1055/s-0038-1676058 · Cited by 25
Sekar, P., Heydarian, H. C., Cnota, J. F., Hornberger, L. K., & Michelfelder, E. C. (2015) · Cardiology in the Young · doi:10.1017/s1047951113001467 · Cited by 23
Zhao, Y., Edington, S., Fleenor, J., Sinkovskaya, E. S., Porche, L. M., & Abuhamad, A. (2016) · Ultrasound in Obstetrics and Gynecology · doi:10.1002/uog.15998 · Cited by 18
Self-citationYeo, L., Markush, D., & Romero, R. J. (2018) · The Journal of Maternal-Fetal & Neonatal Medicine · doi:10.1080/14767058.2018.1484088 · Cited by 17
Evaluation of fetal cardiac contractility by two‐dimensional ultrasonography
Goldinfeld, M., Weiner, E., Peleg, D., Shalev, E., & Ben‐Ami, M. (2004) · Prenatal Diagnosis · doi:10.1002/pd.880 · Cited by 16
Valve Interventions in Utero: Understanding the Timing, Indications, and Approaches
Sizarov, A., & Boudjemline, Y. (2017) · Canadian Journal of Cardiology · doi:10.1016/j.cjca.2017.06.009 · Cited by 16
Cohen, L., Mangers, K., Grobman, W. A., Gotteiner, N. L., Julien, S. D., Dungan, J. S., Fonseca, L., & PLATT, L. D. (2010) · Journal of Ultrasound in Medicine · doi:10.7863/jum.2010.29.5.751 · Cited by 10
Toward a clearer picture of health
Arnaout, R. (2018) · Nature Medicine · doi:10.1038/s41591-018-0318-x · Cited by 8
Self-citationA review of the diagnostic accuracy of fetal cardiac anomalies
Miceli, F. (2015) · Australasian Journal of Ultrasound in Medicine · doi:10.1002/j.2205-0140.2015.tb00024.x · Cited by 6
Flagged first, then by citations
Cited by
Sorted by citations
Machine Learning and the Future of Cardiovascular Care
Quer, G., Arnaout, R., Henne, M., & Arnaout, R. (2021) · Journal of the American College of Cardiology · doi:10.1016/j.jacc.2020.11.030 · Cited by 353
Self-citationVision Transformers in medical computer vision—A contemplative retrospection
Parvaiz, A., Khalid, M. A., Zafar, R., Ameer, H., Ali, M. U., & Fraz, M. M. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106126 · Cited by 302
Transforming Cardiovascular Care With Artificial Intelligence: From Discovery to Practice
Khera, R., Oikonomou, E. K., Nadkarni, G. N., Morley, J., Wiens, J., Butte, A. J., & Topol, E. J. (2024) · Journal of the American College of Cardiology · doi:10.1016/j.jacc.2024.05.003 · Cited by 324
A review on deep-learning algorithms for fetal ultrasound-image analysis
Fiorentino, M. C., Villani, F. P., Cosmo, M. D., Frontoni, E., & Moccia, S. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102629 · Cited by 224
Artificial intelligence in cardiovascular diseases: diagnostic and therapeutic perspectives
Sun, X., Yin, Y., Yang, Q., & Huo, T. (2023) · European journal of medical research · doi:10.1186/s40001-023-01065-y · Cited by 180
Metadata
- Authors
- Arnaout, R., Curran, L., Zhao, Y., Levine, J. C., Chinn, E., & Moon-Grady, A. J.
- Affiliations
- University of California, San Francisco (US)
- Boston Children's Hospital (US)
- Corresponding author
- Arnaout, Rima
- Venue
- Nature Medicine
- Publisher
- openRxiv
- Published
- 2020-06-24
- Volume
- 27
- Issue
- 5
- Pages
- 882-891
- ISSN
- 1078-8956 (ISSN-L)
- PMC
- PMC8380434
- OpenAlex
- W3161999894
- Cited by
- 315
- References
- 51
- Citations by year
- 2026: 55 · 2025: 61 · 2024: 66 · 2023: 60 · 2022: 45 · 2021: 26 · 2020: 2
- FWCI
- 14.24
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Locations
- Nature Medicine (publishedVersion)
- medRxiv (acceptedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- Funders
- U.S. Department of Defense (W81XWH1910294)
- U.S. Department of Health and Human Services (R01HL150394)
- American Heart Association (17IGMV33870001)
- National Institutes of Health (TR991872, UL1 TR991872)
- University of California, San Francisco
- National Heart, Lung, and Blood Institute (R01HL150394)
- National Center for Advancing Translational Sciences (UL1 TR991872)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Cardiology and Cardiovascular Medicine › Congenital Heart Disease Studies
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- complex congenital heart disease, prenatal diagnosis, fetal echocardiography, fetal ultrasound screening, neural network ensembles, sensitivity and specificity
- MeSH terms
- Adult, Biometry, Female, Fetus, Heart, Heart Defects, Congenital, Humans, Mass Screening, Myocardium, Pregnancy, Pregnancy Trimester, Second, Prenatal Diagnosis, Sensitivity and Specificity, Thorax, Ultrasonography, Prenatal, Neural Networks, Computer, Echocardiography, Three-Dimensional, Young Adult
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
