Artificial intelligence in evaluating the fetal heart. The first stages of development of the Moscow Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology

Pathology of the fetal cardiovascular system is the most common type of congenital malformations and is in second place among the causes of infant mortality and accounts for 47% of all causes of death from malformations.

The effectiveness and result of cardiac surgery largely depend on the earliest diagnosis of heart disease, on the readiness of medical staff to provide medical care to a newborn with CHD and on the ability to arrange timely transportation of a newborn to a cardiac surgery center as soon as possible.

Fetal heart assessment is a difficult task, mainly due to the small size of the heart, involuntary fetal movements, and lack of experience in fetal echocardiography by some ultrasound specialists

The objective of our study is to create a medical decision support system by forming an algorithm for examining the fetal heart using AI. The result of which should be one of the medical opinion options: “normal” – correct heart structure – no congenital heart disease; “not normal” – abnormal heart structure – the presence of congenital heart disease cannot be ruled out, extended fetal echocardiography is recommended as soon as possible.

One of the tasks of our work was: the formation of an algorithm for examining the fetal heart using AI, the result of which should be one of the options for a medical conclusion: “norm” – correct heart structure – there is no CHD; “not norm” – incorrect heart structure – the presence of CHD cannot be excluded, extended fetal echocardiography is recommended as soon as possible.

Research methodology: The study was conducted at the gestation period of 18–21 weeks. Each study per patient contained video files of five standard projections of the heart. Each slice is represented by at least 25 frames. Verification was performed by confirming/changing the diagnosis by an expert doctor, as well as confirming the diagnosis after birth.

As a result of the work, the task of determining the zones of the chest and heart of the fetus was solved with an accuracy of 98%; the task of classifying the slice of the heart on the frame was solved with an accuracy of 82%, the task of determining pathology on the slices of the heart was solved with an accuracy of 77%.

Conclusions: The results showed that the artificial intelligence algorithm can improve the accuracy of ultrasound diagnosis of the fetal heart and has good applied value. It is expected that artificial intelligence methods will contribute to the standardization and optimization of fetal echocardiography, increase the percentage of prenatal diagnosis of CHD, and thereby lead to a decrease in infant mortality and childhood disability.

1. Bokeria L.A., Shatalova K.V. Pediatric cardiac surgery. Manual for physicians. Moscow: A.N. Bakulev National Medical Research Center of Cardiovascular Surgery, 2016: 24–40. (In Russian)

2. Bokerija E.L. Perinatal cardiology: the present and the future. Part I: Congenital heart disease. Rossiyskiy Vestnik Perinatologii i Pediatrii = Russian Bulletin of Perinatology and Pediatrics. 2019; 64 (3): 5–10. https://doi.org/10.21508/1027-4065-2019-64-3-5-10 (In Russian)

3. Mcleod G., Shum K., Gupta T. et al. Echocardiography in Congenital Heart Disease. Prog. Cardiovasc. Dis. 2018; 5–6: 468–475. https://doi.org/10.1016/j.pcad.2018.11.004

4. Yoon S.A., Hong W.H., Cho H.J. Congenital heart disease diagnosed with echocardiogram in newborns with asymptomatic cardiac murmurs: A systematic review. BMC Pediatr. 2020; 20: 1–10. https://doi.org/10.1186/s12887-020-02212-8

5. Patel N., Narasimhan E., Kennedy A. Fetal Cardiac US: Techniques and Normal Anatomy Correlated with Adult CT and MR Imaging. Radiographics. 2017; 37 (4): 1290–1303. https://doi.org/10.1148/rg.2017160126

6. Carvalho J.S., Axt-Fliedner R., Chaoui R. et al. ISUOG Practice Guidelines (updated): fetal cardiac screening. Ultrasound Obstet. Gynecol. 2023. 61 (6): 788–803. https://doi.org/10.1002/uog.26224

7. Order of the Ministry of Health of the Russian Federation of October 20, 2020 N 1130n "On approval of the Procedure for the provision of medical care in the profile of "obstetrics and gynecology": [Registered with the Ministry of Justice of the Russian Federation on November 12, 2020. Registration N 60869]. - Official Internet portal of legal information" (www.pravo.gov.ru) November 13, 2020 N 0001202011130037 (date of access: 04/13/2025). - Text: electronic. (In Russian)

8. Carvalho J.S., Allan L.D., Chaoui R. et al. ISUOG Practice Guidelines (updated): sonographic screening examination of the fetal heart. Ultrasound Obstet. Gynecol. 2013; 41 (3): 348–359. https://doi.org/10.1002/uog.12403

9. Bak G.S., Shaffer B.L., Madriago E. et al. Impact of maternal obesity on fetal cardiac screening: which follow-up strategy is cost-effective? Ultrasound Obstet. Gynecol. 2020; 56 (5): 705–716. https://doi.org/10.1002/uog.21895

10. Friedberg M.K., Silverman N.H., Moon-Grady A.J. et al. Prenatal detection of congenital heart disease. J. Pediatr. 2009; 155 (1): 26–31, 31.e1. https://doi.org/10.1016/j.jpeds.2009.01.050

11. Sekar P., Heydarian H.C., Cnota J.F. et al. Diagnosis of congenital heart disease in an era of universal prenatal ultrasound screening in southwest Ohio. Cardiol. Young. 2015; 25 (1): 35–41. https://doi.org/10.1017/S1047951113001467

12. Donofrio M.T., Moon-Grady A.J., Hornberger L.K. et al. Diagnosis and treatment of fetal cardiac disease: a scientific statement from the American Heart Association. American Heart Association Adults With Congenital Heart Disease Joint Committee of the Council on Cardiovascular Disease in the Young and Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and Council on Cardiovascular and Stroke Nursing. Circulation. 2014; 27; 129 (21): 2183–2242. https://doi.org/10.1161/01.cir.0000437597.44550.5d

13. Tunçalp Ӧ., Pena-Rosas J.P., Lawrie T. et al. WHO recommendations on antenatal care for a positive pregnancy experience – going beyond survival. BJOG. 2017; 124 (6): 860–862. https://doi.org/10.1161/01.cir.0000437597.44550.5d

14. Baumgartner C.F., Kamnitsas K., Matthew J. et al. SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound. IEEE Trans. Med. Imaging. 2017; 36 (11): 2204–2215. https://doi.org/10.1109/TMI.2017.2712367

15. Sklansky M., DeVore G.R. Fetal Cardiac Screening: What Are We (and Our Guidelines) Doing Wrong? J. Ultrasound Med. 2016; 35 (4): 679–681. https://doi.org/10.7863/ultra.15.07021

16. Sun H.Y., Proudfoot J.A., McCandless R.T. Prenatal detection of critical cardiac outflow tract anomalies remains suboptimal despite revised obstetrical imaging guidelines. Congenit. Heart Dis. 2018; 13 (5): 748–756. https://doi.org/10.1111/chd.12648

17. Garcia-Canadilla P., Sanchez-Martinez S., Crispi F. et al. Machine Learning in Fetal Cardiology: What to Expect. Fetal Diagn. Ther. 2020; 47 (5): 363–372. https://doi.org/10.1159/000505021

18. Arnaout R., Curran L., Zhao Y. et al. An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease. Nat. Med. 2021; 27 (5): 882–891. https://doi.org/10.1038/s41591-021-01342-5

19. Artificial intelligence technologies. United Kingdom Engineering and Physical Sciences Research Council. Accessed December 12, 2022. https://epsrc.ukri.org/research/ourportfolio/researchareas/ait/

20. He F., Wang Y., Xiu Y.et al. Artificial intelligence in prenatal ultrasound diagnosis. Front. Med. 2021; 8: 729978. https://doi.org/10.3389/fmed.2021.729978

21. Zhang L., Ye X., Lambrou T. et al. A supervised texton based approach for automatic segmentation and measurement of the fetal head and femur in 2D ultrasound images. Phys. Med. Biol. 2016; 61 (3): 1095–1115. https://doi.org/10.1088/ 0031-9155/61/3/1095

22. Li P., Zhao H., Liu P., Cao F. Automated measurement network for accurate segmentation and parameter modification in fetal head ultrasound images. Med. Biol. Eng. Comput. 2020; 58 (11): 2879–2892. https://doi.org/10.1007/s11517-020-02242-5

23. van den Heuvel T.L.A., Petros H., Santini S. et al. Automated fetal head detection and circumference estimation from free-hand ultrasound sweeps using deep learning in resource-limited countries. Ultrasound Med. Biol. 2019; 45 (3): 773–785. https://doi.org/10.1016/j.ultrasmedbio.2018.09.015

24. Sinclair M., Baumgartner C.F., Matthew J. et al. Human-Level Performance on Automatic Head Biometrics in Fetal Ultrasound Using Fully Convolutional Neural Networks. IEEE. 2018; 714–717. https://doi.org/10.48550/arXiv.1804.09102

25. Nurmaini S., Rachmatullah M.N., Sapitri A.I. et al. Deep Learning-Based Computer-Aided Fetal Echocardiography: Application to Heart Standard View Segmentation for Congenital Heart Defects Detection. Sensors (Basel). 2021; 21 (23): 8007. https://doi.org/10.3390/s21238007

26. Han G., Jin T., Zhang L. et al. Adoption of Compound Echocardiography under Artificial Intelligence Algorithm in Fetal Congenial Heart Disease Screening during Gestation. Bionics Biomech. 2022: 2022: 6410103. https://doi.org/10.1155/2022/6410103

27. Wu H., Wu B., Lai F. et al. Application of Artificial Intelligence in Anatomical Structure Recognition of Standard Section of Fetal Heart. Comput. Math. Methods. Med. 2023. 24; 2023: 5650378. https://doi.org/10.1155/2023/5650378