Learning deep architectures for the interpretation of first-trimester fetal echocardiography (LIFE) - a study protocol for developing an automated intelligent decision support system for early fetal echocardiography.
Artificial intelligence
Decision support systems
Echocardiography
Fetal cardiology
Journal
BMC pregnancy and childbirth
ISSN: 1471-2393
Titre abrégé: BMC Pregnancy Childbirth
Pays: England
ID NLM: 100967799
Informations de publication
Date de publication:
11 Jan 2023
11 Jan 2023
Historique:
received:
07
04
2022
accepted:
09
11
2022
entrez:
11
1
2023
pubmed:
12
1
2023
medline:
14
1
2023
Statut:
epublish
Résumé
Congenital Heart Disease represents the most frequent fetal malformation. The lack of prenatal identification of congenital heart defects can have adverse consequences for the neonate, while a correct prenatal diagnosis of specific cardiac anomalies improves neonatal care neurologic and surgery outcomes. Sonographers perform prenatal diagnosis manually during the first or second-trimester scan, but the reported detection rates are low. This project's primary objective is to develop an Intelligent Decision Support System that uses two-dimensional video files of cardiac sweeps obtained during the standard first-trimester fetal echocardiography (FE) to signal the presence/absence of previously learned key features. The cross-sectional study will be divided into a training part of the machine learning approaches and the testing phase on previously unseen frames and eventually on actual video scans. Pregnant women in their 12-13 + 6 weeks of gestation admitted for routine first-trimester anomaly scan will be consecutively included in a two-year study, depending on the availability of the experienced sonographers in early fetal cardiac imaging involved in this research. The Data Science / IT department (DSIT) will process the key planes identified by the sonographers in the two- dimensional heart cine loop sweeps: four-chamber view, left and right ventricular outflow tracts, three vessels, and trachea view. The frames will be grouped into the classes representing the plane views, and then different state-of-the- art deep-learning (DL) pre-trained algorithms will be tested on the data set. The sonographers will validate all the intermediary findings at the frame level and the meaningfulness of the video labeling. FE is feasible and efficient during the first trimester. Still, the continuous training process is impaired by the lack of specialists or their limited availability. Therefore, in our study design, the sonographer benefits from a second opinion provided by the developed software, which may be very helpful, especially if a more experienced colleague is unavailable. In addition, the software may be implemented on the ultrasound device so that the process could take place during the live examination. The study is registered under the name "Learning deep architectures for the Interpretation of Fetal Echocardiography (LIFE)", project number 408PED/2020, project code PN-III-P2-2.1-PED-2019. ClinicalTrials.gov , unique identifying number NCT05090306, date of registration 30.10.2020.
Sections du résumé
BACKGROUND
BACKGROUND
Congenital Heart Disease represents the most frequent fetal malformation. The lack of prenatal identification of congenital heart defects can have adverse consequences for the neonate, while a correct prenatal diagnosis of specific cardiac anomalies improves neonatal care neurologic and surgery outcomes. Sonographers perform prenatal diagnosis manually during the first or second-trimester scan, but the reported detection rates are low. This project's primary objective is to develop an Intelligent Decision Support System that uses two-dimensional video files of cardiac sweeps obtained during the standard first-trimester fetal echocardiography (FE) to signal the presence/absence of previously learned key features.
METHODS
METHODS
The cross-sectional study will be divided into a training part of the machine learning approaches and the testing phase on previously unseen frames and eventually on actual video scans. Pregnant women in their 12-13 + 6 weeks of gestation admitted for routine first-trimester anomaly scan will be consecutively included in a two-year study, depending on the availability of the experienced sonographers in early fetal cardiac imaging involved in this research. The Data Science / IT department (DSIT) will process the key planes identified by the sonographers in the two- dimensional heart cine loop sweeps: four-chamber view, left and right ventricular outflow tracts, three vessels, and trachea view. The frames will be grouped into the classes representing the plane views, and then different state-of-the- art deep-learning (DL) pre-trained algorithms will be tested on the data set. The sonographers will validate all the intermediary findings at the frame level and the meaningfulness of the video labeling.
DISCUSSION
CONCLUSIONS
FE is feasible and efficient during the first trimester. Still, the continuous training process is impaired by the lack of specialists or their limited availability. Therefore, in our study design, the sonographer benefits from a second opinion provided by the developed software, which may be very helpful, especially if a more experienced colleague is unavailable. In addition, the software may be implemented on the ultrasound device so that the process could take place during the live examination.
TRIAL REGISTRATION
BACKGROUND
The study is registered under the name "Learning deep architectures for the Interpretation of Fetal Echocardiography (LIFE)", project number 408PED/2020, project code PN-III-P2-2.1-PED-2019.
TRIAL REGISTRATION
BACKGROUND
ClinicalTrials.gov , unique identifying number NCT05090306, date of registration 30.10.2020.
Identifiants
pubmed: 36631859
doi: 10.1186/s12884-022-05204-x
pii: 10.1186/s12884-022-05204-x
pmc: PMC9832772
doi:
Banques de données
ClinicalTrials.gov
['NCT05090306']
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
20Subventions
Organisme : Ministerul Educației și Cercetării Științifice
ID : 408PED/2020
Commentaires et corrections
Type : ErratumIn
Informations de copyright
© 2023. The Author(s).
Références
Hoffman JI. Incidence of congenital heart disease: II. Prenatal incidence. Pediatr Cardiol. 1995;16(4):155–65.
pubmed: 7567659
doi: 10.1007/BF00794186
Crispi F, Gratacós E. Fetal cardiac function: technical considerations and potential research and clinical applications. Fetal Diagn Ther. 2012;32(1–2):47–64.
pubmed: 22614129
doi: 10.1159/000338003
Gardiner HM. Advances in fetal echocardiography. Semin Fetal Neonatal Med. 2018;23(2):112–8.
pubmed: 29221765
doi: 10.1016/j.siny.2017.11.006
Tibballs J, Cantwell-Bartl A. Outcomes of management decisions by parents for their infants with hypoplastic left heart syndrome born with and without a prenatal diagnosis. J Paediatr Child Health. 2008;44(6):321–4.
pubmed: 18194197
doi: 10.1111/j.1440-1754.2007.01265.x
Franklin O, Burch M, Manning N, et al. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. Heart. 2002;87(1):67–9.
pubmed: 11751670
pmcid: 1766965
doi: 10.1136/heart.87.1.67
Acherman RJ, Evans WN, Luna CF, et al. Prenatal detection of congenital heart disease in southern Nevada: the need for universal fetal cardiac evaluation. J Ultrasound Med. 2007;26(12):1715–9.
pubmed: 18029923
doi: 10.7863/jum.2007.26.12.1715
Schultz AH, Localio AR, Clark BJ, et al. Epidemiologic features of the presentation of critical congenital heart disease: implications for screening. Pediatrics. 2008;121(4):751–7.
pubmed: 18381540
doi: 10.1542/peds.2007-0421
Brown KL, Ridout DA, Hoskote A, et al. Delayed diagnosis of congenital heart disease worsens preoperative condition and outcome of surgery in neonates. Heart. 2006;92(9):1298–302.
pubmed: 16449514
pmcid: 1861169
doi: 10.1136/hrt.2005.078097
Tworetzky W, McElhinney DB, Reddy VM, et al. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Circulation. 2001;103(9):1269–73.
pubmed: 11238272
doi: 10.1161/01.CIR.103.9.1269
Verheijen PM, Lisowski LA, Stoutenbeek P, et al. Lactacidosis in the neonate is minimized by prenatal detection of congenital heart disease. Ultrasound Obstet Gynecol. 2002;19:552–5.
pubmed: 12047532
doi: 10.1046/j.1469-0705.2002.00714.x
Kumar RK, Newburger JW, Gauvreau K, et al. Comparison of outcome when hypoplastic left heart syndrome and transposition of the great arteries are diagnosed prenatally versus when diagnosis of these two conditions is made only postnatally. Am J Cardiol. 1999;83(12):1649–53.
pubmed: 10392870
doi: 10.1016/S0002-9149(99)00172-1
Mahle WT, Clancy RR, McGaurn SP, et al. Impact of prenatal diagnosis on survival and early neurologic morbidity in neonates with the hypoplastic left heart syndrome. Pediatrics. 2001;107(6):1277–82.
pubmed: 11389243
doi: 10.1542/peds.107.6.1277
Bonnet D, Coltri A, Butera G, et al. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Circulation. 1999;99(7):916–8.
pubmed: 10027815
doi: 10.1161/01.CIR.99.7.916
Khoshnood B, De Vigan C, Vodovar V, et al. Trends in prenatal diagnosis, pregnancy termination, and perinatal mortality of newborns with congenital heart disease in France, 1983-2000: a population-based evaluation. Pediatrics. 2005;115(1):95–101.
pubmed: 15629987
doi: 10.1542/peds.2004-0516
Bensemlali M, Stirnemann J, Le Bidois J, et al. Discordances between prenatal and postnatal diagnoses of congenital heart diseases and impact on care strategies. J Am Coll Cardiol. 2016;68(9):921–30.
pubmed: 27561766
doi: 10.1016/j.jacc.2016.05.087
Mozumdar N, Rowland J, Pan S, et al. Diagnostic accuracy of fetal echocardiography in congenital heart disease. J Am Soc Echocardiogr. 2020;33(11):1384–90.
pubmed: 32828627
doi: 10.1016/j.echo.2020.06.017
Abu-Rustum RS, Ziade MF, Abu-Rustum SE. Learning curve and factors influencing the feasibility of performing fetal echocardiography at the time of the first-trimester scan. J Ultrasound Med. 2011;30(5):695–700.
pubmed: 21527618
doi: 10.7863/jum.2011.30.5.695
Nemescu D, Onofriescu M. Factors affecting the feasibility of routine first-trimester fetal echocardiography. J Ultrasound Med. 2015;34(1):161–6.
pubmed: 25542952
doi: 10.7863/ultra.34.1.161
DeVore GR, Medearis AL, Bear MB, et al. Fetal echocardiography: factors that influence imaging of the fetal heart during the second trimester of pregnancy. J Ultrasound Med. 1993;12(11):659–63.
pubmed: 8264018
doi: 10.7863/jum.1993.12.11.659
Paladini D. Sonography in obese and overweight pregnant women: clinical, medicolegal and technical issues. Ultrasound Obstet Gynecol. 2009;33(6):720–9.
pubmed: 19479683
doi: 10.1002/uog.6393
Tegnander E, Eik-Nes SH. The examiner's ultrasound experience significantly impacts the detection rate of congenital heart defects at the second-trimester fetal examination. Ultrasound Obstet Gynecol. 2006;28(1):8–14.
pubmed: 16736449
doi: 10.1002/uog.2804
Hunter S, Heads A, Wyllie J, et al. Prenatal diagnosis of congenital heart disease in the northern region of England: benefits of a training program for obstetric ultrasonographers. Heart. 2000;84(3):294–8.
pubmed: 10956294
pmcid: 1760944
doi: 10.1136/heart.84.3.294
International Society of Ultrasound in Obstetrics and Gynecology Guidelines. Cardiac screening examination of the fetus: guidelines for performing the "basic" and "extended basic" cardiac scans. Ultrasound Obstet Gynecol. 2006;27:107–13.
doi: 10.1002/uog.2677
Iliescu DG. Echocardiography. USA: SM Group; 2016.
Bennasar M, Martinez JM, Olivella A, del Rio M, Gomez O, Figueras F, et al. Feasibility and accuracy of fetal echocardiography using four-dimensional spatiotemporal image correlation technology before 16 weeks gestation. Ultrasound Obstet Gynecol. 2009;33:645–51.
pubmed: 19479815
doi: 10.1002/uog.6374
Viñals F, Mandujano L, Vargas G, Giuliano A. Prenatal diagnosis of congenital heart disease using four-dimensional Spatio-temporal image correlation (STIC) telemedicine via an internet link a pilot study. Ultrasound Obstet Gynecol. 2005;25:25–31.
pubmed: 15593355
doi: 10.1002/uog.1796
Bennasar M, Martínez JM, Gómez O, Figueras F, Olivella A, Puerto B, et al. Intra- and interobserver repeatability of fetal cardiac examination using four-dimensional spatiotemporal image correlation in each trimester of pregnancy. Ultrasound Obstet Gynecol. 2010;35:318–23.
pubmed: 20127758
doi: 10.1002/uog.7570
Turan S, Turan OM, Ty-Torredes K, Harman CR, Baschat AA. Standardization of the first-trimester fetal cardiac examination using spatiotemporal image correlation with tomographic ultrasound and color Doppler imaging. Ultrasound Obstet Gynecol. 2009;33:652–6.
pubmed: 19405042
doi: 10.1002/uog.6372
Tudorache S, Cara M, Iliescu DG, Novac L, Cernea N. First trimester two- and four-dimensional cardiac scan: intra-and interobserver agreement, comparison between methods and benefits of color Doppler technique. Ultrasound Obstet Gynecol. 2013;42(6):659–68.
pubmed: 23494803
doi: 10.1002/uog.12459
Iliescu D, Tudorache S, Comanescu A, Antsaklis P, Cotarcea S, Novac L, et al. Improved detection rate of structural abnormalities in the first trimester using an extended examination protocol. Ultrasound Obstet Gynecol. 2013;42(3):300–9.
pubmed: 23595897
doi: 10.1002/uog.12489
Belciug S, Gorunescu F. Intelligent decision support systems—a journey to smarter healthcare. Switzerland: Springer Nature; 2020.
doi: 10.1007/978-3-030-14354-1
Garcia-Canadilla P, Sanchez-Martinez S, Crispi F, et al. Machine learning in fetal cardiology: what to expect. Fetal Diagn Ther. 2020;47(5):363–72.
pubmed: 31910421
doi: 10.1159/000505021
Matsuoka R, Komatsu M, Sakai A, et al. A novel deep learning-based system for fetal cardiac screening. Ultrasound Obstet Gynecol. 2019;56(1):177–8.
doi: 10.1002/uog.20945
Komatsu R, Matsuoka R, Arakaki T, et al. Novel AI-guided ultrasound screening system for fetal heart can demonstrate findings in timeline diagram. Ultrasound Obstet Gynecol. 2019;54(1):134.
doi: 10.1002/uog.20796
Arnaout R, Curran L, Chinn E, et al. Deep-learning models improve community-level diagnosis for common congenital heart disease lesions. 2019.
Stoean R, Iliescu D, Stoean C, Ilie V, Patru C, Hotoleanu M, et al. Deep learning for the detection of frames of interest in fetal heart assessment from first trimester ultrasound, 16th international work-conference on artificial neural networks, June, 16th–18th; 2021.
Carvalho JS, Allan LD, Chaoui R, et al. ISUOG practice guidelines (updated): sonographic screening examination of the fetal heart. Ultrasound Obstet Gynecol. 2013;41:348–59.
pubmed: 23460196
doi: 10.1002/uog.12403
Becker R, Wegner RD. Detailed screening for fetal anomalies and cardiac defects at the 11–13-week scan. Ultrasound Obstet Gynecol. 2006;27:613–8.
pubmed: 16570262
doi: 10.1002/uog.2709
Lombardi CM, Bellotti M, Fesslova V, Cappellini A. Fetal echocardiography at the time of the nuchal translucency scan. Ultrasound Obstet Gynecol. 2007;29:249–57.
Persico N, Moratalla J, Lombardi CM, Zidere V, Allan L, Nicolaides KH. Fetal echocardiography at 11–13 weeks by transabdominal high-frequency ultrasound. Ultrasound Obstet Gynecol. 2011;37:296–301.
pubmed: 21229572
doi: 10.1002/uog.8934
Selvaraju RR, Cogswell M, Das A, et al. Grad-CAM: visual explanations from deep networks via gradient-based localization, 2017 IEEE International Conference on Computer Vision (ICCV); 2017. p. 618–26.
Ter Haar G, Duck FA, Shaw A, et al. The safe use of ultrasound in medical diagnosis. UK: The British Institute of Radiology; 2012.
doi: 10.1259/bir.2012.0001
Bioeffects and Safety Committee, Salvesen K, Lees C, Abramowicz J, et al. ISUOG-WFUMB statement on the non- medical use of ultrasound, 2011. Ultrasound Obstet Gynecol. 2011;38(5):608.
doi: 10.1002/uog.10107
Campbell S, Platt L. The publishing of papers on first-trimester Doppler. Ultrasound Obstet Gynecol. 1999;14:159–60.
pubmed: 10550872
doi: 10.1046/j.1469-0705.1999.14030159.x
Salvesen K, Lees C, Abramowicz J, Brezinka C, Ter Haar G, Maršál K, et al. ISUOG statement on the safe use of Doppler in the 11 to 13 + 6- week fetal ultrasound examination. Ultrasound Obstet Gynecol. 2011;37:628.
pubmed: 21618313
doi: 10.1002/uog.9026
Salvesen KA, Lees C, Abramowicz J, Brezinka C, Ter Haar G, Maršál K. Safe use of Doppler ultrasound during the 11 to 13 + 6-week scan: is it possible? Ultrasound Obstet Gynecol. 2011;37:625–8.
pubmed: 21618312
doi: 10.1002/uog.9025