Machine learning to predict outcomes of fetal cardiac disease: a pilot study

Machine learning to predict outcomes of fetal cardiac disease: a pilot study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Machine learning to predict outcomes of fetal cardiac disease: a pilot study Lynne Nield, Cedric Manlhiot, Keziah Magor, Lindsay Freud, Bhargava Chinni, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4045996/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 May, 2024 Read the published version in Pediatric Cardiology → Version 1 posted 7 You are reading this latest preprint version Abstract BACKGROUND: Prediction of outcomes following a prenatal diagnosis of congenital heart disease is challenging. Machine learning (ML) algorithms may be used to reduce clinical uncertainty and improve prognostic accuracy. METHODS: We performed a pilot study to train ML algorithms to predict postnatal outcomes based on clinical data. Specific objectives were to predict 1) in-utero or neonatal death, 2) high-acuity neonatal care and 3) favourable outcomes. We included all fetuses with cardiac disease at Sunnybrook Health Sciences Centre, Toronto, Canada, from 2012 – 2021. Prediction models were created using the XgBoost algorithm (tree-based) with 5-fold cross validation. RESULTS: Among 211 cases of fetal cardiac disease, 61 were excluded (39 terminations, 21 lost to follow-up, 1 isolated arrhythmia), leaving a cohort of 150 fetuses. Fifteen (10%) demised (10 neonates) and 70 (52%) of live births required high acuity neonatal care. Of those with clinical follow-up, 57/82 (70%) had a favourable outcome. Prediction models for live birth, high acuity neonatal care and favourable outcome had AUCs of 0.75, 0.82 and 0.72, respectively. The most important predictors for death were the presence of non-cardiac or genetic abnormalities and more severe structural heart disease. High acuity of postnatal care was predicted by increased nuchal thickness, lower gestational age (GA) and birthweight and favourable outcome was predicted by normal fetal right ventricular function, no tricuspid valve abnormalities, and normal GA/weight at birth. CONCLUSION Prediction models using ML provide good discrimination of key prenatal and postnatal outcomes among fetuses with congenital heart disease. Machine Learning fetal cardiology congenital heart disease outcomes Figures Figure 1 Figure 2 Introduction Predictive analytics, in other words data-driven predictions generated from a group of patients and applied to a specific case, are increasingly being used by healthcare providers to reduce clinical uncertainty, improve their prognostic accuracy and potentially improve patient management and outcomes [1, 2]. This would be particularly useful in the field of fetal cardiology where outcome prediction for the individual fetus remains difficult [3-6]. Indeed, upon the diagnosis of a cardiac lesion, patients are typically counselled by a multi-disciplinary team, including fetal cardiologists, maternal-fetal medicine specialists, and neonatal intensive care physicians. This counselling results in parents being provided with a range of possible clinical outcomes which can vary between full surgical repair, no neonatal intervention or single ventricle palliation with associated risks of significant morbidity and death [7-14]. The ambiguity in predicting clinical outcomes creates challenges for families trying to not only grapple with a complicated cardiac diagnosis but to make decisions regarding pregnancy termination vs. continuation vs. palliative compassionate care. Machine learning technology is emerging as a new and exciting method to offset this uncertainty and is already being used in a variety of settings[15-17]. A recent publication from Dr. Moon-Grady’s group in San Francisco showed that neural networks can be trained to identify normal and abnormal fetal hearts. The next step is to use a dataset comprised of clinical factors and image information to predict the progression and postnatal outcomes of fetuses with specific forms of congenital heart disease (CHD)[18]. This retrospective pilot study aimed to investigate utilizing machine learning (ML) algorithms to create predictive models for salient prenatal and postnatal outcomes for fetuses with congenital heart disease. Methods Patient population This retrospective study included all fetuses diagnosed with CHD from January 2012 to December 2021 at a single tertiary centre in Canada (Sunnybrook Health Sciences Centre). The hospital performs approximately 20,000 prenatal ultrasounds and up to 500 fetal echocardiograms per year. Peri- and postnatal outcomes were collected from Sunnybrook, but also from the affiliated obstetric and neonatal centers (Mount Sinai Hospital, Toronto, Michael Garron Hospital, Toronto and The Hospital for Sick Children, Toronto) depending on the ultimate location of delivery and postnatal care. The study was approved by the Research Ethics Board of all participating institutions as well as and Johns Hopkins University where the analysis was performed. The requirement for individual patient consent was waived for a retrospective study. Clinical characteristics and neonatal outcomes ( Table 1 ) were collected through chart review. The severity of congenital heart disease was defined according to the Hoffman criteria as mild, moderate or severe (See Appendix A ) [19]. Machine learning algorithms were then developed to predict the following outcomes of interest: 1) in utero demise/stillbirth or death within 72 hours of birth despite planned active care, 2) need for high level neonatal care (delivery at a tertiary care hospital, prostaglandins, neonatal intensive care or intensive care admission, mechanical ventilation, neonatal surgical or catheter intervention < 30 days of life) and 3) favourable postnatal outcomes defined as survival without severe developmental delay at last follow up, which was extracted from the patient’s chart. The severity of congenital heart disease was defined according to the Hoffman criteria as mild, moderate or severe (See Appendix A ) [19]. Predictive features and clinical outcomes The feature set consisted of 70 potential predictors; 62 out of 70 predictors were integrated in all three models including information about demographics, comorbidities, medical management, and fetal structural findings from the fetal echocardiogram comprising cardiac anatomy. Additionally, 7 more predictors including labor induction, mode of delivery, sex, gestational age at birth, birth weight and Apgar score were used in the models predicting the need for high acuity neonatal care and favourable outcomes (69 predictors total for these 2 models). Finally, the ML model predicting the risk of adverse outcomes also included postnatal cardiac intervention (surgical or catheter based) information in addition to the 69 variables listed above. Data preprocessing Missing values imputation We generated an analysis dataset for every ML model, comprising a subset of patients with exclusively recorded outcome values. Three separate analysis datasets were constructed, each aligning with the corresponding outcome and its associated number of predictors. To address missing information within each dataset, we employed a predictive imputation method [20]. This method considers the similarity between patients in each dataset. An iterative imputation algorithm was implemented, allowing up to 50 cycles. In each cycle, a decision tree regressor was applied to each dataset, aiding in discerning patterns among patients and relationships between predictors to approximate the missing measurements. After estimating missing values in the three analysis datasets, predictor variables with more than two categories underwent transformation using one-hot encoding. Tree based machine learning model induction and evaluation The XGBoost tree-based ML algorithm [21] was applied to each of these datasets. The use of the XGBoost algorithm facilitated the categorization of patients into two distinct groups, allowing for the assessment of non-linear relationships between predictors and their respective outcomes. To improve the XGBoost predictions, optimization was performed using the area under the receiver-operating characteristic curve (AUC) as a benchmark to evaluate model effectiveness. Furthermore, the XGBoost algorithm underwent hyperparameter tuning [22] to achieve optimal results. This tuning process involved 5-fold cross-validation (CV), utilizing Bayesian optimization techniques [23] and implementing a search grid to identify the combination of XGBoost parameters that maximized the area under the curve (AUC). We have employed SHAP (SHapley Additive exPlanations) method to gain insights into influence of individual features on the model's predictions [25]. SHAP values were calculated for each predictor across all patients, and we illustrated the impact of each feature on the model’s log-odds prediction through a beeswarm plot. Features with higher SHAP values contribute more significantly to the model's decision-making process, and are displayed further away from the center regardless of whether they increase or decrease the predicted outcome. In estimated the 95% confidence interval (CI) for the AUC metric, bootstrapping was employed with 500 resamples per fold across the 5-fold CV, yielding a cumulative total of 2500 bootstraps for each model. The CI was then determined utilizing the standard error derived from the distribution of bootstrapped AUC values. All the analyses were implemented using Python version 3.9.12. Results Clinical outcomes Between January 1, 2012 and December 31, 2021 a total of 1576 fetal echocardiograms were performed, of which 211 (13%) fetuses were diagnosed with congenital heart disease. Sixty-one cases (29%) were excluded due to pregnancy termination (N=39), loss to follow up (N=21) and benign arrhythmia (N=1) (isolated premature atrial contractions with structurally normal heart). This left a total cohort of 150 fetuses for analysis. At the diagnostic fetal echocardiogram (mean 24 6/7 weeks gestation), there were 63 (37%) cases with minor cardiac abnormalities and 46 (31%) with major cardiac abnormalities. In another 41 (27%) the fetuses had an initial normal fetal echocardiogram but later had milder forms of CHD at prenatal follow up. Non-cardiac abnormalities were seen in 24/111 (22% cases) and genetic abnormalities were present in 19/63 (30% of those tested prenatally). There were 15 (10%) perinatal deaths (5 in utero, 10 neonatal). Among the 135 live births, 70 (52%) neonates needed high acuity neonatal care. Of the liveborn patients with follow up, 57/82 (70%) were alive at last follow up without severe developmental delays. Table 1 depicts the summary of maternal and fetal characteristics stratified by the need for high acuity neonatal care. Performance of prediction models Figure 1 depictes the area under the receiver operating characteristic (ROC) curves for the three XGBoost ML models. Prediction models for fetal or neonatal death, high acuity neonatal care and an favourable outcome had AUC’s of 0.75 (range 0.742 to 0.758), 0.82 (range 0.814 to 0.826) and 0.72 (0.717 to 0.723), respectively. Performance metrics obtained from the 5-fold cross-validation were aggregated as presented in Table 2 . The ROC curves ( Figure 1) and associated AUC values collectively indicate that the three XGBoost models have effectively balanced sensitivity and specificity as displayed in Table 2 . This equilibrium between true positive and false positive rates suggests that the inducted models have found a reasonable compromise between correctly identifying positive instances and minimizing false alarms. The ML model for high acuity neonatal care performed better relative to the other two models and may be attributed to the higher prevalence of outcomes (ie. balanced data distribution of the two patient classes). Figures 2A, 2B and 2C depict the most salient features for each of the prediction models, which were derived utilizing absolute SHAP values. Features informing absence of non-cardiac abnormalities, higher maternal age, nuchal thickness size and indication of no previous births contributed significantly towards the prediction of fetal or neonatal death. High acuity neonatal care was impacted by lower values of Apgar at 1 minute, birth weight, maternal anti-Ro positive antibody, gestational age at birth, and presence of more severe structural CHD, and older maternal age. A favourable outcome was predicted by the absence of genetic abnormalities and higher values of gestational age at birth and a normal right ventricle and/or tricuspid valve . Discussion This pilot study created a ML model to predict fetal or neonatal death, the need for high acuity neonatal care and a favourable clinical outcome in fetuses with CHD. The strongest predictors of outcome were extra-cardiac abnormality, an underlying genetic diagnosis and severity of CHD. The use of the SHAP (SHapley Additive exPlanations) method is a specific ML modality that identifies the variables that contribute to the ML decision-making algorithm in a binary manner, creating a model where the variable can increase or decrease the predicted outcome [24]. This differs from a regression analysis which is a more linear statistical finding and predicts a specific outcome (positive or negative) as opposed to either direction. This type of ML model has the potential to become more accurate with increased numbers of cases. This would allow a formal, tailored prediction model for families that would give improved prognostication and specific actuarial risks of adverse outcomes. Such a model could expand to include additional relevant outcomes and morbidities, such as the need for repeated surgical or catheter interventions, exercise intolerance, arrhythmias or cardiac failure. A future ML model could include diagnostic testing data such as anatomic ultrasound and fetal echocardiogram images [25-27]. Serial fetal echocardiograms from first trimester to delivery also have the potential to create even more accurate ML models to determine which fetuses will have progressive changes[28, 29]. A favourable outcome was predicted in our model by the variables of fetal right ventricular function and tricuspid valve abnormalities. This has been noted previously as a risk factor for in-utero demise in fetuses with CHD [30, 31] . The physiology of the fetal heart is right ventricle dominant. Given that a fetal heart can only increase heart rate to improve cardiac output unlike the neonate who can increase stroke volume, the interplay between the right heart and fetal well-being is very delicate [32-34]. Abnormal right ventricular function and tricuspid valve abnormalities were not predictors of in utero or neonatal death in our algorithm. However, this may be due to the small size of the study. This study validated the feasibility of developing ML models for fetuses with CHD but would require a large, multicenter prospective patient database to create a truly functional model for individual patients. Creating a real-time ML model for clinicians would improve accuracy of prenatal counselling both expectant parents and health care providers alike. LIMITATIONS The study has the limitations of a retrospective study. First, we had 21 patients who were lost to follow. We would presume, however, that those cases did not have critical cardiac disease, otherwise they most certainly would have been evaluated at the Hospital for Sick Children for surgical or interventional management. Second, the type of cardiac lesions was very heterogeneous. We sought to offset this limitation by categorizing them into mild vs. severe diseases. Finally, the fetal echocardiogram images were not used in the machine learning model for this study, and we only included image interpretation from final reports. This facilitated the analysis as it overcomes potential errors in image recognition. We opted for the 5-fold CV technique to assess the ML models over the train-test split criterion. This choice was driven by two main considerations. First, given our limited patient population, CV allowed for more effective use of the data, resulting in a more thorough evaluation of the ML model's generalization performance. Second, some of the analysis datasets exhibited a significant imbalance in outcomes, and employing CV could help alleviate this imbalance by ensuring that each fold represents a balanced distribution of both patient classes. Conclusion Prediction models using ML provide good discrimination of key prenatal and postnatal outcomes among fetuses with CHD. A prospective, multi-center registry to gather more robust data has the potential to provide the clinician with clearer information in order to more accurately counsel families with a fetal diagnosis of CHD. Abbreviations Congenital heart disease = CHD Machine Learning = ML CV = cross-validation Declarations There are no conflicts of interest to disclose. This research study was not funded. References Oberije, C., et al., A prospective study comparing the predictions of doctors versus models for treatment outcome of lung cancer patients: a step toward individualized care and shared decision making. Radiother Oncol, 2014. 112 (1): p. 37-43. Hatch, S., Uncertainty in medicine. BMJ, 2017. 357 : p. j2180. Donofrio, M.T., et al., Diagnosis and treatment of fetal cardiac disease: a scientific statement from the American Heart Association. Circulation, 2014. 129 (21): p. 2183-242. Pinto, N.M., et al., Prenatal cardiac care: Goals, priorities & gaps in knowledge in fetal cardiovascular disease: Perspectives of the Fetal Heart Society. Prog Pediatr Cardiol, 2020. 59 : p. 101312. Carvalho, J.S., et al., Clinical impact of first and early second trimester fetal echocardiography on high risk pregnancies. Heart, 2004. 90 (8): p. 921-6. Yu, D., L. Sui, and N. Zhang, Performance of First-Trimester Fetal Echocardiography in Diagnosing Fetal Heart Defects: Meta-analysis and Systematic Review. J Ultrasound Med, 2020. 39 (3): p. 471-480. Morgan, C.T., et al., Improving Prenatal Diagnosis of Coarctation of the Aorta. Can J Cardiol, 2019. 35 (4): p. 453-461. Freud, L.R., et al., Fetal aortic valvuloplasty for evolving hypoplastic left heart syndrome: postnatal outcomes of the first 100 patients. Circulation, 2014. 130 (8): p. 638-45. MacColl, C.E., et al., Risk factors associated with in utero demise in fetuses with congenital heart disease: A case/control study. Cardiology in the Young, 2012. 22 : p. S74. Allan, L.D., H.D. Apfel, and B.F. Printz, Outcome after prenatal diagnosis of the hypoplastic left heart syndrome. Heart, 1998. 79 (4): p. 371-3. Sharland, G.K., K.Y. Chan, and L.D. Allan, Coarctation of the aorta: Difficulties in prenatal diagnosis. British Heart Journal, 1994. 71 (1): p. 70-75. Weber, R.W., et al., Foetal echocardiographic assessment of borderline small left ventricles can predict the need for postnatal intervention. Cardiology in the Young, 2013. 23 (1): p. 99-107. O'Brien, S.M., et al., An empirically based tool for analyzing mortality associated with congenital heart surgery. J Thorac Cardiovasc Surg, 2009. 138 (5): p. 1139-53. Brown, D.W., et al., Prenatal diagnosis of congenital heart disease and association with morbidity among survivors of initial palliation for single ventricle heart disease: Analysis of the national pediatric cardiology quality improvement collaborative database. Circulation, 2011. 1) . Bernard, O., et al., Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved? IEEE Trans Med Imaging, 2018. 37 (11): p. 2514-2525. LeCun, Y., Y. Bengio, and G. Hinton, Deep learning. Nature, 2015. 521 (7553): p. 436-44. Ouyang, D., et al., Video-based AI for beat-to-beat assessment of cardiac function. Nature, 2020. 580 (7802): p. 252-256. Arnaout, R., et al., An ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease. Nat Med, 2021. 27 (5): p. 882-891. Hoffman, J.I. and S. Kaplan, The incidence of congenital heart disease. J Am Coll Cardiol, 2002. 39 (12): p. 1890-900. Fouad, K.M., et al., Advanced methods for missing values imputation based on similarity learning. PeerJ Comput Sci, 2021. 7 : p. e619. Chen, T. and C. Guestrin, XGBoost: A Scalable Tree Boosting System , in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . 2016, Association for Computing Machinery: San Francisco, California, USA. p. 785–794. Bergstra, J. and Y. Bengio, Random search for hyper-parameter optimization. J. Mach. Learn. Res., 2012. 13 (null): p. 281–305. Snoek, J., H. Larochelle, and R.P. Adams, Practical Bayesian Optimization of Machine Learning Algorithms. Advances in neural information processing systems (NIPS), 2012: p. 2951-2959. Lundberg, S.M., et al., From Local Explanations to Global Understanding with Explainable AI for Trees. Nat Mach Intell, 2020. 2 (1): p. 56-67. Athalye, C. and R. Arnaout, Domain-guided data augmentation for deep learning on medical imaging. PLoS One, 2023. 18 (3): p. e0282532. Athalye, C., et al., Deep learning model for prenatal congenital heart disease (CHD) screening can be applied to retrospective imaging from the community setting, outperforming initial clinical detection in a well-annotated cohort. Ultrasound Obstet Gynecol, 2023. Truong, V.T., et al., Application of machine learning in screening for congenital heart diseases using fetal echocardiography. Int J Cardiovasc Imaging, 2022. Gardiner, H.M., First-trimester fetal echocardiography: routine practice or research tool? Ultrasound Obstet Gynecol, 2013. 42 (6): p. 611-2. Zidere, V., et al., Comparison of echocardiographic findings in fetuses at less than 15 weeks' gestation with later cardiac evaluation. Ultrasound Obstet Gynecol, 2013. 42 (6): p. 679-86. MacColl, C.E., et al., Factors associated with in utero demise of fetuses that have underlying cardiac pathologies. Pediatr Cardiol, 2014. 35 (8): p. 1403-14. Jepson, B.M., et al., Pregnancy loss in major fetal congenital heart disease: incidence, risk factors and timing. Ultrasound Obstet Gynecol, 2023. 62 (1): p. 75-87. Rudolph, A.M., Circulatory changes during gestational development of the sheep and human fetus. Pediatr Res, 2018. 84 (3): p. 348-351. Rudolph, A.M. and M.A. Heymann, The fetal circulation. Annu Rev Med, 1968. 19 : p. 195-206. Sun, L., et al., Understanding Fetal Hemodynamics Using Cardiovascular Magnetic Resonance Imaging. Fetal Diagn Ther, 2020. 47 (5): p. 354-362. Tables TABLE 1: Maternal and fetal characteristics stratified by high acuity neonatal care N High acuity neonatal care N No high acuity neonatal care N All patients p Maternal Factors Maternal age (years) 70 35.1±5.2 65 33.6±4.8 135 34.4±5.1 0.11 Previous livebirths 63 0.8±1.2 60 0.9±1.0 123 0.8±1.1 0.78 Twin pregnancy 68 7 (10.3%) 65 1 (1.5%) 133 8 (6.0%) 0.06 Maternal diabetes (all types) 69 10 (14.5%) 65 3 (4.6%) 134 13 (9.7%) 0.07 Anti-Ro positive antibody 69 4 (5.8%) 64 27 (42.4%) 133 31 (23.3%) <.001 Fertility treatments (IVF or IUI) 69 11 (19.0%) 65 8 (14.0%) 134 19 (14.2%) 0.62 Fetal Factors Increased NT (3.5 mm and above) 40 9 (22.5%) 28 4 (14.3%) 68 13 (19.1%) 0.54 Genetic Abnormalities 35 12 (34.3%) 28 7 (25.0%) 63 19 (30.2%) 0.58 Non-cardiac abnormalities 56 15 (26.8%) 55 9 (16.4%) 111 24 (21.6%) 0.25 Fetal growth restriction 60 12 (20.0%) 62 3 (4.8%) 122 15 (12.3%) 0.01 Placental abnormalities 56 8 (14.3%) 55 1 (1.8%) 111 9 (8.1%) 0.03 Severity of structural CHD (major) 60 39 (65.0%) 36 15 (41.7%) 96 54 (56.3%) 0.03 Hoffman cardiac severity > mild 54 19 (35.2%) 31 18 (58.1%) 85 37 (43.5%) 0.07 Cardiac Diagnoses Right heart disease 70 29 (41.4%) 65 17 (26.2%) 135 46 (34.1%) 0.07 Left heart disease 70 9 (12.9%) 65 0 (0%) 135 9 (6.7%) 0.003 Atrioventricular septal defect 69 7 (10.1%) 64 1 (1.6%) 133 8 (6.0%) 0.06 Right ventricle hypoplasia 69 17 (24.6%) 64 8 (12.5%) 133 25 (18.8%) 0.08 Abnormal right ventricle function 69 9 (13%) 64 1 (1.6%) 133 10 (7.5%) 0.02 Left ventricle hypoplasia 70 13 (18.6%) 65 4 (6.2%) 135 17 (12.6%) 0.04 Abnormal tricuspid valve 69 14 (20.3%) 64 7 (10.9%) 133 21 (15.8%) 0.16 Abnormal pulmonary valve 70 7 (10.0%) 64 3 (4.7%) 134 10 (7.5%) 0.33 Abnormal aortic valve 70 6 (8.6%) 65 1 (1.5%) 135 7 (5.2%) 0.12 Abnormal mitral valve 69 4 (5.8%) 65 2 (3.1%) 134 6 (4.5%) 0.68 Pericardial Effusion 70 10 (14.3%) 65 3 (4.6%) 135 13 (9.6%) 0.08 Postnatal Outcome Sex (male) 63 28 (44.4%) 57 27 (47.4%) 120 55 (45.8%) 0.85 Gestational age at birth (weeks) 69 35.3±5.0 57 38.0±2.2 126 36.5±4.2 <.001 Birth weight (kg) 60 2.4±1.0 54 3.0±0.5 114 2.7±0.9 <.001 Apgar 1 min 50 7.1±2.4 44 8.6±1.0 94 7.8±2.0 <.001 Apgar 5 min 50 8.2±1.7 44 8.9±0.4 94 8.5±1.3 0.01 Cardiac intervention (postnatal) 64 27 (42.2%) 45 3 (6.7%) 109 30 (27.5%) <.001 SLE, systemic lupus erythematosus; NT, nuchal thickness; GU= genitourinary; ; VSD, ventricular septal defect; SVD, spontaneous vaginal delivery; IVF, in vitro fertilization; IUI, in utero insemination Table 2: Evaluation of the prediction models for fetal or neonatal death, high acuity neonatal care and favourable outcome algorithms. PPV and NPV represent positive and negative predictive values, AUC represent the area under the receiver-operating characteristic curve. Outcome Number of patients Outcome prevalence Number of features in the model AUC Sensitivity Specificity PPV NPV Perinatal death 150 0.90 62 0.75 0.63 0.67 0.94 0.20 High acuity neonatal care 135 0.52 69 0.80 0.73 0.75 0.76 0.72 Favourable outcome 87 0.31 70 0.72 0.63 0.57 0.40 0.77 Additional Declarations No competing interests reported. Supplementary Files AppendixA.docx Cite Share Download PDF Status: Published Journal Publication published 09 May, 2024 Read the published version in Pediatric Cardiology → Version 1 posted Editorial decision: Revision requested 04 Apr, 2024 Reviews received at journal 02 Apr, 2024 Reviewers agreed at journal 12 Mar, 2024 Reviewers invited by journal 12 Mar, 2024 Editor assigned by journal 10 Mar, 2024 Submission checks completed at journal 10 Mar, 2024 First submitted to journal 08 Mar, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4045996","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":278913057,"identity":"c3cd42d3-3f2b-48e1-98bd-e9a106665ea2","order_by":0,"name":"Lynne Nield","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYBACPijNw8DOYPgAzJQAEQdwa2GDa2FmMDYgSQsDUIuZBHFaJHKPPfjAcEeGn5l5W+XPtjt5Brd7Hz74cIZBnr8Bl5a8dMMZDM94JJvZym7ztj0rNrhz3Nhwxg0Gwxk4bGKTyDGT5mE4zGNwmMfsNmPb4cQNN9LYpHk+MCTgchxYyx+gFnuglsKfyFrk8WlhANnCzGPGwAvXcoMhwQCXFp43ZpI9QFdJHGYrluY5dzhx5p1jzIYzzkgYbsShhZ89x0ziR8Vhe/725o0ff5QdTuy73cb44MMxG3k5HFoYBBKAhAGSgAJEpQQO9SBr0M2Sb8CteBSMglEwCkYmAACcqFhT0RN7QwAAAABJRU5ErkJggg==","orcid":"","institution":"Sunnybrook Health Sciences Centre","correspondingAuthor":true,"prefix":"","firstName":"Lynne","middleName":"","lastName":"Nield","suffix":""},{"id":278913058,"identity":"45b2fe76-95f7-493e-a15e-e9ff113ef9d8","order_by":1,"name":"Cedric Manlhiot","email":"","orcid":"","institution":"Johns Hopkins University Baltimore MD","correspondingAuthor":false,"prefix":"","firstName":"Cedric","middleName":"","lastName":"Manlhiot","suffix":""},{"id":278913059,"identity":"0cea8456-289f-4d54-8cc6-da2245a54125","order_by":2,"name":"Keziah Magor","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Keziah","middleName":"","lastName":"Magor","suffix":""},{"id":278913061,"identity":"4f7d443b-3ab2-43a8-86cf-5b0f8cb0bca8","order_by":3,"name":"Lindsay Freud","email":"","orcid":"","institution":"The Hospital for Sick Children","correspondingAuthor":false,"prefix":"","firstName":"Lindsay","middleName":"","lastName":"Freud","suffix":""},{"id":278913062,"identity":"e2260f57-c417-4fc4-90a6-4552b9e33772","order_by":4,"name":"Bhargava Chinni","email":"","orcid":"","institution":"Johns Hopkins University Baltimore MD","correspondingAuthor":false,"prefix":"","firstName":"Bhargava","middleName":"","lastName":"Chinni","suffix":""},{"id":278913064,"identity":"d89d5bd2-c7bf-4b5c-9eac-1327fa76718b","order_by":5,"name":"Andrew Ims","email":"","orcid":"","institution":"Johns Hopkins University Baltimore MD","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Ims","suffix":""},{"id":278913065,"identity":"e4c2ed52-d1d3-494e-855a-d4138ed17f67","order_by":6,"name":"Nir Melamed","email":"","orcid":"","institution":"Sunnybrook Health Sciences Centre","correspondingAuthor":false,"prefix":"","firstName":"Nir","middleName":"","lastName":"Melamed","suffix":""},{"id":278913067,"identity":"1a2f2820-506e-4bf3-b70b-617a0c20817a","order_by":7,"name":"Ori Nevo","email":"","orcid":"","institution":"Sunnybrook Health Sciences Centre","correspondingAuthor":false,"prefix":"","firstName":"Ori","middleName":"","lastName":"Nevo","suffix":""},{"id":278913068,"identity":"4805918b-c555-40c9-8018-a13c984f7cff","order_by":8,"name":"Dany Weisz","email":"","orcid":"","institution":"Sunnybrook Health Sciences Centre","correspondingAuthor":false,"prefix":"","firstName":"Dany","middleName":"","lastName":"Weisz","suffix":""},{"id":278913069,"identity":"1bbcc68f-d3fa-421f-9023-9ed2b6c50c01","order_by":9,"name":"Stefania Ronzoni","email":"","orcid":"","institution":"Sunnybrook Health Sciences Centre","correspondingAuthor":false,"prefix":"","firstName":"Stefania","middleName":"","lastName":"Ronzoni","suffix":""}],"badges":[],"createdAt":"2024-03-08 15:06:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4045996/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4045996/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00246-024-03512-x","type":"published","date":"2024-05-09T21:17:48+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":52619905,"identity":"4903eeba-a150-4c39-baa2-f75fd432d34f","added_by":"auto","created_at":"2024-03-13 16:41:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":738666,"visible":true,"origin":"","legend":"\u003cp\u003eArea under the receiver-operating characteristic curve (AUC)\u003cstrong\u003e \u003c/strong\u003efor the three XGBoost models: predicting fetal or neonatal death (orange, AUC = 0.75), high acuity neonatal care (blue, AUC = 0.80) and a favourable outcome (red, AUC = 0.72)\u003c/p\u003e","description":"","filename":"MLstudy.Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4045996/v1/1c9e58cdf19158e96832abc9.png"},{"id":52620739,"identity":"7b083dcc-fe6b-45f7-9e7c-7eeb242a7038","added_by":"auto","created_at":"2024-03-13 16:49:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1039165,"visible":true,"origin":"","legend":"\u003cp\u003eFeature importance obtained using SHAP (SHapley Additive exPlanations) values represented with a beeswarm chart. Each point represents a feature value and inform the relationships between the features and model’s log-odds prediction. Beeswarm charts illustrating the absolute SHAP values, showcasing features that contribute to 80% of the total feature importance associated with prediction models estimating the risk of patient experiencing \u003cstrong\u003e(A) \u003c/strong\u003efetal or neonatal death, \u003cstrong\u003e(B)\u003c/strong\u003e high acuity neonatal care and \u003cstrong\u003e(C)\u003c/strong\u003e favourable outcome.\u003c/p\u003e","description":"","filename":"MLstudy.Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-4045996/v1/53d277305f39b243ae0a9bf1.png"},{"id":56488102,"identity":"80cf76db-fa81-44f3-8459-2cd88a097eb2","added_by":"auto","created_at":"2024-05-14 21:29:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":819390,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4045996/v1/05a1c4c4-3eb6-4851-8466-6c6614678872.pdf"},{"id":52619903,"identity":"d87e7e76-927a-442c-9093-4965dc43e55e","added_by":"auto","created_at":"2024-03-13 16:41:38","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":15584,"visible":true,"origin":"","legend":"","description":"","filename":"AppendixA.docx","url":"https://assets-eu.researchsquare.com/files/rs-4045996/v1/8caeebd0e0e71148dfe69caa.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine learning to predict outcomes of fetal cardiac disease: a pilot study","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePredictive analytics, in other words data-driven predictions generated from a group of patients and applied to a specific case, are increasingly being used by healthcare providers to reduce clinical uncertainty, improve their prognostic accuracy and potentially improve patient management and outcomes\u0026nbsp;[1, 2]. \u0026nbsp; This would be particularly useful in the field of fetal cardiology where outcome prediction for the individual fetus remains difficult\u0026nbsp;[3-6]. \u0026nbsp; Indeed, upon the diagnosis of a cardiac lesion, patients are typically counselled by a multi-disciplinary team, including fetal cardiologists, maternal-fetal medicine specialists, and neonatal intensive care physicians. \u0026nbsp;This counselling results in parents being provided with a range of possible clinical outcomes which can vary between full surgical repair, no neonatal intervention or single ventricle palliation with associated risks of significant morbidity and death\u0026nbsp;[7-14]. \u0026nbsp; \u0026nbsp;The ambiguity in predicting clinical outcomes creates challenges for families trying to not only grapple with a complicated cardiac diagnosis but to make decisions regarding pregnancy termination vs. continuation vs. palliative compassionate care.\u003c/p\u003e\n\u003cp\u003eMachine learning technology is emerging as a new and exciting method to offset this uncertainty and is already being used in a variety of settings[15-17]. \u0026nbsp;A recent publication from Dr. Moon-Grady’s group in San Francisco showed that neural networks can be trained to identify normal and abnormal fetal hearts. \u0026nbsp;The next step is to use a dataset comprised of clinical factors and image information to predict the progression and postnatal outcomes of fetuses with specific forms of congenital heart disease (CHD)[18]. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis retrospective pilot study aimed to investigate utilizing machine learning (ML) algorithms to create predictive models for salient prenatal and postnatal outcomes for fetuses with congenital heart disease.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003ePatient population\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective study included all fetuses diagnosed with CHD from January 2012 to December 2021 at a single tertiary centre in Canada (Sunnybrook Health Sciences Centre). The hospital performs approximately 20,000 prenatal ultrasounds and up to 500 fetal echocardiograms per year. \u0026nbsp;Peri- and postnatal outcomes were collected from Sunnybrook, but also from the affiliated obstetric and neonatal centers (Mount Sinai Hospital, Toronto, Michael Garron Hospital, Toronto and The Hospital for Sick Children, Toronto) depending on the ultimate location of delivery and postnatal care. \u0026nbsp; The study was approved by the Research Ethics Board of all participating institutions as well as and Johns Hopkins University where the analysis was performed. \u0026nbsp;The requirement for individual patient consent was waived for a retrospective study.\u003c/p\u003e\n\u003cp\u003eClinical characteristics and \u0026nbsp;neonatal outcomes (\u003cstrong\u003eTable 1\u003c/strong\u003e) \u0026nbsp;were collected through chart review.\u0026nbsp; \u0026nbsp;The severity of congenital heart disease was defined according to the Hoffman criteria as mild, moderate or severe \u003cstrong\u003e(See Appendix A\u003c/strong\u003e)\u0026nbsp;[19].\u003c/p\u003e\n\u003cp\u003eMachine learning algorithms were then developed to predict the following\u0026nbsp;outcomes of interest:\u0026nbsp;1) in utero demise/stillbirth or death within 72 hours of birth despite planned active care, 2) need for high level neonatal care (delivery at a tertiary care hospital, prostaglandins, neonatal intensive care or intensive care admission, mechanical ventilation, neonatal surgical or catheter intervention \u0026lt; 30 days of life) and 3) favourable postnatal outcomes defined as\u0026nbsp;survival without severe developmental delay at last follow up, which was extracted from the patient’s chart. \u0026nbsp;The severity of congenital heart disease was defined according to the Hoffman criteria as mild, moderate or severe \u003cstrong\u003e(See Appendix A\u003c/strong\u003e)\u0026nbsp;[19].\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePredictive features and clinical outcomes\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;The feature set consisted of 70 potential predictors; 62 out of 70 predictors were integrated in all three models including information about demographics, comorbidities, medical management, and fetal structural findings from the fetal echocardiogram comprising cardiac anatomy. Additionally, 7 more predictors including labor induction, mode of delivery, sex, gestational age at birth, birth weight and Apgar score were used in the models predicting the need for high acuity neonatal care and favourable outcomes (69 predictors total for these 2 models). Finally, the ML model predicting the risk of adverse outcomes also included postnatal cardiac intervention (surgical or catheter based) information in addition to the 69 variables listed above. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData preprocessing\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMissing values imputation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe generated an analysis dataset for every ML model, comprising a subset of patients with exclusively recorded outcome values. Three separate analysis datasets were constructed, each aligning with the corresponding outcome and its associated number of predictors. To address missing information within each dataset, we employed a predictive imputation method\u0026nbsp;[20]. This method considers the similarity between patients in each dataset. An iterative imputation algorithm was implemented, allowing up to 50 cycles. In each cycle, a decision tree regressor was applied to each dataset, aiding in discerning patterns among patients and relationships between predictors to approximate the missing measurements.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter estimating missing values in the three analysis datasets, predictor variables with more than two categories underwent transformation using one-hot encoding.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTree based machine learning model induction and evaluation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe XGBoost tree-based ML algorithm\u0026nbsp;[21]\u0026nbsp;was applied to each of these datasets. The use of the XGBoost algorithm facilitated the categorization of patients into two distinct groups, allowing for the assessment of \u003cu\u003enon-linear\u003c/u\u003e relationships between predictors and their respective outcomes. To improve the XGBoost predictions, optimization was performed using the area under the receiver-operating characteristic curve (AUC) as a benchmark to evaluate model effectiveness. Furthermore, the XGBoost algorithm underwent hyperparameter tuning\u0026nbsp;[22]\u0026nbsp;to achieve optimal results. This tuning process involved 5-fold cross-validation (CV), utilizing Bayesian optimization techniques\u0026nbsp;[23]\u0026nbsp;and implementing a search grid to identify the combination of XGBoost parameters that maximized the area under the curve (AUC).\u003c/p\u003e\n\u003cp\u003eWe have employed SHAP (SHapley Additive exPlanations) method to gain insights into influence of individual features on the model's predictions [25]. SHAP values were calculated for each predictor across all patients, and we illustrated the impact of each feature on the model’s log-odds prediction through a beeswarm plot. Features with higher SHAP values contribute more significantly to the model's decision-making process, and are displayed further away from the center regardless of whether they increase or decrease the predicted outcome. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn estimated the 95% confidence interval (CI) for the AUC metric, bootstrapping was employed with 500 resamples per fold across the 5-fold CV, yielding a cumulative total of 2500 bootstraps for each model. \u0026nbsp;The CI was then determined utilizing the standard error derived from the distribution of bootstrapped AUC values. \u0026nbsp; All the analyses were implemented using Python version 3.9.12.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003eClinical outcomes\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBetween January 1, 2012 and December 31, 2021 a total of 1576 fetal echocardiograms were performed, of which 211 (13%) fetuses were diagnosed with congenital heart disease. \u0026nbsp;Sixty-one cases (29%) were excluded due to pregnancy termination (N=39), loss to follow up (N=21) and benign arrhythmia (N=1) (isolated premature atrial contractions with structurally normal heart). \u0026nbsp;This left a total cohort of 150 fetuses for analysis. \u0026nbsp;At the diagnostic fetal echocardiogram (mean 24 6/7 weeks gestation), there were 63 (37%) cases with minor cardiac abnormalities and 46 (31%) with major cardiac abnormalities. \u0026nbsp;In another 41 (27%) the fetuses had an initial normal fetal echocardiogram but later had milder forms of CHD at prenatal follow up. Non-cardiac abnormalities were seen in 24/111 (22% cases) and genetic abnormalities were present in 19/63 (30% of those tested prenatally).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThere were 15 (10%) perinatal deaths (5 in utero, 10 neonatal). \u0026nbsp;\u0026nbsp;Among the 135 live births, 70 (52%) neonates \u0026nbsp;needed\u0026nbsp;high acuity neonatal care.\u0026nbsp;Of the liveborn patients with follow up, 57/82 (70%) were alive at last follow up without severe developmental delays. \u0026nbsp;\u003cstrong\u003eTable 1\u003c/strong\u003e depicts the summary of maternal and fetal characteristics stratified by the need for high acuity neonatal care. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePerformance of prediction models\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 1\u003c/strong\u003e depictes the area under the receiver operating characteristic (ROC) curves for the three XGBoost ML models.\u0026nbsp;Prediction models for\u0026nbsp;fetal or neonatal death, high acuity neonatal care and an favourable outcome\u0026nbsp;had AUC’s of 0.75 (range 0.742 to 0.758), 0.82 (range 0.814 to 0.826) and 0.72 (0.717 to 0.723), \u0026nbsp;respectively. \u0026nbsp;Performance metrics obtained from the 5-fold cross-validation were aggregated as presented in \u003cstrong\u003eTable 2\u003c/strong\u003e. The ROC curves (\u003cstrong\u003eFigure 1)\u003c/strong\u003e and associated AUC values collectively indicate that the three XGBoost models have effectively balanced sensitivity and specificity as displayed in \u003cstrong\u003eTable 2\u003c/strong\u003e. This equilibrium between true positive and false positive rates suggests that the inducted models have found a reasonable compromise between correctly identifying positive instances and minimizing false alarms. The ML model for high acuity neonatal care performed better relative to the other two models and may be attributed to the higher prevalence of outcomes (ie. balanced data distribution of the two patient classes). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigures 2A, 2B and 2C\u003c/strong\u003e depict the most salient features for each of the prediction models, \u0026nbsp;which were derived utilizing absolute SHAP values. \u0026nbsp;Features informing absence of non-cardiac abnormalities, higher maternal age, nuchal thickness size and indication of no previous births contributed significantly towards the prediction of fetal or neonatal death. \u0026nbsp;High acuity neonatal care was impacted by lower values of Apgar at 1 minute, birth weight, maternal anti-Ro positive antibody, gestational age at birth, and presence of more severe structural CHD, and older maternal age. \u0026nbsp;A favourable outcome was predicted by the absence \u0026nbsp;of genetic abnormalities and higher values of gestational age at birth and a normal right ventricle and/or tricuspid valve . \u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis pilot study created a ML model to predict fetal or neonatal death, the need for high acuity neonatal care and a favourable clinical outcome in fetuses with CHD. \u0026nbsp;The strongest predictors of outcome were extra-cardiac abnormality, an underlying genetic diagnosis and severity of CHD. \u0026nbsp;The use of the\u0026nbsp;SHAP (SHapley Additive exPlanations) method is a specific ML modality that identifies the variables that contribute to the ML decision-making algorithm in a binary manner, creating a model where the variable can increase or decrease the predicted outcome\u0026nbsp;[24]. \u0026nbsp;This differs from a regression analysis which is a more linear statistical finding and predicts a specific outcome (positive or negative) as opposed to either direction. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis type of ML model has the potential to become more accurate with increased numbers of cases. \u0026nbsp;This would allow a formal, tailored prediction model for families that would give improved prognostication and specific actuarial risks of adverse outcomes. \u0026nbsp; Such a model could expand to include additional relevant outcomes and morbidities, such as the need for repeated surgical or catheter interventions, exercise intolerance, arrhythmias or cardiac failure. \u0026nbsp;A future ML model could include diagnostic testing data such as anatomic ultrasound and fetal echocardiogram images\u0026nbsp;[25-27]. \u0026nbsp; Serial fetal echocardiograms from first trimester to delivery also have the potential to create even more accurate ML models to determine which fetuses will have progressive changes[28, 29].\u003c/p\u003e\n\u003cp\u003eA favourable outcome was predicted in our model by the variables of fetal right ventricular function and tricuspid valve abnormalities. \u0026nbsp;This has been noted previously as a risk factor for in-utero demise in fetuses with CHD\u0026nbsp;[30, 31]\u0026nbsp;. \u0026nbsp; The physiology of the fetal heart is right ventricle dominant. \u0026nbsp;Given that a fetal heart can only increase heart rate to improve cardiac output unlike the neonate who can increase stroke volume, the interplay between the right heart and fetal well-being is very delicate\u0026nbsp;[32-34]. \u0026nbsp;Abnormal right ventricular function and tricuspid valve abnormalities were not predictors of in utero or neonatal death in our algorithm. However, this may be due to the small size of the study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study validated the feasibility of developing ML models for fetuses with CHD but would require a large, multicenter prospective patient database to create a truly functional model for individual patients. \u0026nbsp;Creating a real-time ML model for clinicians would improve accuracy of prenatal counselling both expectant parents and health care providers alike.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLIMITATIONS\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe study has the limitations of a retrospective study. \u0026nbsp;First, we had 21 patients who were lost to follow. \u0026nbsp; We would presume, however, that those cases did not have critical cardiac disease, otherwise they most certainly would have been evaluated at the Hospital for Sick Children for surgical or interventional management. \u0026nbsp;Second, the type of cardiac lesions was very heterogeneous. \u0026nbsp;We sought to offset this limitation by categorizing them into mild vs. severe diseases. \u0026nbsp; Finally, the fetal echocardiogram images were not used in the machine learning model for this study, and we only included image interpretation from final reports. \u0026nbsp;This facilitated the analysis as it overcomes potential errors in image recognition. \u0026nbsp;We opted for the 5-fold CV technique to assess the ML models over the train-test split criterion. This choice was driven by two main considerations. First, given our limited patient population, CV allowed for more effective use of the data, resulting in a more thorough evaluation of the ML model's generalization performance. Second, some of the analysis datasets exhibited a significant imbalance in outcomes, and employing CV could help alleviate this imbalance by ensuring that each fold represents a balanced distribution of both patient classes.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003ePrediction models using ML provide good discrimination of key prenatal and postnatal outcomes among fetuses with CHD. \u0026nbsp;A prospective, multi-center registry to gather more robust data has the potential to provide the clinician with clearer information in order to more accurately counsel families with a fetal diagnosis of CHD.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCongenital heart disease = CHD\u003c/p\u003e\n\u003cp\u003eMachine Learning = ML\u003c/p\u003e\n\u003cp\u003eCV = cross-validation\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eThere are no conflicts of interest to disclose.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis research study was not funded.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eOberije, C., et al., \u003cem\u003eA prospective study comparing the predictions of doctors versus models for treatment outcome of lung cancer patients: a step toward individualized care and shared decision making.\u003c/em\u003e Radiother Oncol, 2014. \u003cstrong\u003e112\u003c/strong\u003e(1): p. 37-43.\u003c/li\u003e\n\u003cli\u003eHatch, S., \u003cem\u003eUncertainty in medicine.\u003c/em\u003e BMJ, 2017. \u003cstrong\u003e357\u003c/strong\u003e: p. j2180.\u003c/li\u003e\n\u003cli\u003eDonofrio, M.T., et al., \u003cem\u003eDiagnosis and treatment of fetal cardiac disease: a scientific statement from the American Heart Association.\u003c/em\u003e Circulation, 2014. \u003cstrong\u003e129\u003c/strong\u003e(21): p. 2183-242.\u003c/li\u003e\n\u003cli\u003ePinto, N.M., et al., \u003cem\u003ePrenatal cardiac care: Goals, priorities \u0026amp; gaps in knowledge in fetal cardiovascular disease: Perspectives of the Fetal Heart Society.\u003c/em\u003e Prog Pediatr Cardiol, 2020. \u003cstrong\u003e59\u003c/strong\u003e: p. 101312.\u003c/li\u003e\n\u003cli\u003eCarvalho, J.S., et al., \u003cem\u003eClinical impact of first and early second trimester fetal echocardiography on high risk pregnancies.\u003c/em\u003e Heart, 2004. \u003cstrong\u003e90\u003c/strong\u003e(8): p. 921-6.\u003c/li\u003e\n\u003cli\u003eYu, D., L. Sui, and N. Zhang, \u003cem\u003ePerformance of First-Trimester Fetal Echocardiography in Diagnosing Fetal Heart Defects: Meta-analysis and Systematic Review.\u003c/em\u003e J Ultrasound Med, 2020. \u003cstrong\u003e39\u003c/strong\u003e(3): p. 471-480.\u003c/li\u003e\n\u003cli\u003eMorgan, C.T., et al., \u003cem\u003eImproving Prenatal Diagnosis of Coarctation of the Aorta.\u003c/em\u003e Can J Cardiol, 2019. \u003cstrong\u003e35\u003c/strong\u003e(4): p. 453-461.\u003c/li\u003e\n\u003cli\u003eFreud, L.R., et al., \u003cem\u003eFetal aortic valvuloplasty for evolving hypoplastic left heart syndrome: postnatal outcomes of the first 100 patients.\u003c/em\u003e Circulation, 2014. \u003cstrong\u003e130\u003c/strong\u003e(8): p. 638-45.\u003c/li\u003e\n\u003cli\u003eMacColl, C.E., et al., \u003cem\u003eRisk factors associated with in utero demise in fetuses with congenital heart disease: A case/control study.\u003c/em\u003e Cardiology in the Young, 2012. \u003cstrong\u003e22\u003c/strong\u003e: p. S74.\u003c/li\u003e\n\u003cli\u003eAllan, L.D., H.D. Apfel, and B.F. Printz, \u003cem\u003eOutcome after prenatal diagnosis of the hypoplastic left heart syndrome.\u003c/em\u003e Heart, 1998. \u003cstrong\u003e79\u003c/strong\u003e(4): p. 371-3.\u003c/li\u003e\n\u003cli\u003eSharland, G.K., K.Y. Chan, and L.D. Allan, \u003cem\u003eCoarctation of the aorta: Difficulties in prenatal diagnosis.\u003c/em\u003e British Heart Journal, 1994. \u003cstrong\u003e71\u003c/strong\u003e(1): p. 70-75.\u003c/li\u003e\n\u003cli\u003eWeber, R.W., et al., \u003cem\u003eFoetal echocardiographic assessment of borderline small left ventricles can predict the need for postnatal intervention.\u003c/em\u003e Cardiology in the Young, 2013. \u003cstrong\u003e23\u003c/strong\u003e(1): p. 99-107.\u003c/li\u003e\n\u003cli\u003eO\u0026apos;Brien, S.M., et al., \u003cem\u003eAn empirically based tool for analyzing mortality associated with congenital heart surgery.\u003c/em\u003e J Thorac Cardiovasc Surg, 2009. \u003cstrong\u003e138\u003c/strong\u003e(5): p. 1139-53.\u003c/li\u003e\n\u003cli\u003eBrown, D.W., et al., \u003cem\u003ePrenatal diagnosis of congenital heart disease and association with morbidity among survivors of initial palliation for single ventricle heart disease: Analysis of the national pediatric cardiology quality improvement collaborative database.\u003c/em\u003e Circulation, 2011. \u003cstrong\u003e1)\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eBernard, O., et al., \u003cem\u003eDeep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?\u003c/em\u003e IEEE Trans Med Imaging, 2018. \u003cstrong\u003e37\u003c/strong\u003e(11): p. 2514-2525.\u003c/li\u003e\n\u003cli\u003eLeCun, Y., Y. Bengio, and G. Hinton, \u003cem\u003eDeep learning.\u003c/em\u003e Nature, 2015. \u003cstrong\u003e521\u003c/strong\u003e(7553): p. 436-44.\u003c/li\u003e\n\u003cli\u003eOuyang, D., et al., \u003cem\u003eVideo-based AI for beat-to-beat assessment of cardiac function.\u003c/em\u003e Nature, 2020. \u003cstrong\u003e580\u003c/strong\u003e(7802): p. 252-256.\u003c/li\u003e\n\u003cli\u003eArnaout, R., et al., \u003cem\u003eAn ensemble of neural networks provides expert-level prenatal detection of complex congenital heart disease.\u003c/em\u003e Nat Med, 2021. \u003cstrong\u003e27\u003c/strong\u003e(5): p. 882-891.\u003c/li\u003e\n\u003cli\u003eHoffman, J.I. and S. Kaplan, \u003cem\u003eThe incidence of congenital heart disease.\u003c/em\u003e J Am Coll Cardiol, 2002. \u003cstrong\u003e39\u003c/strong\u003e(12): p. 1890-900.\u003c/li\u003e\n\u003cli\u003eFouad, K.M., et al., \u003cem\u003eAdvanced methods for missing values imputation based on similarity learning.\u003c/em\u003e PeerJ Comput Sci, 2021. \u003cstrong\u003e7\u003c/strong\u003e: p. e619.\u003c/li\u003e\n\u003cli\u003eChen, T. and C. Guestrin, \u003cem\u003eXGBoost: A Scalable Tree Boosting System\u003c/em\u003e, in \u003cem\u003eProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining\u003c/em\u003e. 2016, Association for Computing Machinery: San Francisco, California, USA. p. 785\u0026ndash;794.\u003c/li\u003e\n\u003cli\u003eBergstra, J. and Y. Bengio, \u003cem\u003eRandom search for hyper-parameter optimization.\u003c/em\u003e J. Mach. Learn. Res., 2012. \u003cstrong\u003e13\u003c/strong\u003e(null): p. 281\u0026ndash;305.\u003c/li\u003e\n\u003cli\u003eSnoek, J., H. Larochelle, and R.P. Adams, \u003cem\u003ePractical Bayesian Optimization of Machine Learning Algorithms.\u003c/em\u003e Advances in neural information processing systems (NIPS), 2012: p. 2951-2959.\u003c/li\u003e\n\u003cli\u003eLundberg, S.M., et al., \u003cem\u003eFrom Local Explanations to Global Understanding with Explainable AI for Trees.\u003c/em\u003e Nat Mach Intell, 2020. \u003cstrong\u003e2\u003c/strong\u003e(1): p. 56-67.\u003c/li\u003e\n\u003cli\u003eAthalye, C. and R. Arnaout, \u003cem\u003eDomain-guided data augmentation for deep learning on medical imaging.\u003c/em\u003e PLoS One, 2023. \u003cstrong\u003e18\u003c/strong\u003e(3): p. e0282532.\u003c/li\u003e\n\u003cli\u003eAthalye, C., et al., \u003cem\u003eDeep learning model for prenatal congenital heart disease (CHD) screening can be applied to retrospective imaging from the community setting, outperforming initial clinical detection in a well-annotated cohort.\u003c/em\u003e Ultrasound Obstet Gynecol, 2023.\u003c/li\u003e\n\u003cli\u003eTruong, V.T., et al., \u003cem\u003eApplication of machine learning in screening for congenital heart diseases using fetal echocardiography.\u003c/em\u003e Int J Cardiovasc Imaging, 2022.\u003c/li\u003e\n\u003cli\u003eGardiner, H.M., \u003cem\u003eFirst-trimester fetal echocardiography: routine practice or research tool?\u003c/em\u003e Ultrasound Obstet Gynecol, 2013. \u003cstrong\u003e42\u003c/strong\u003e(6): p. 611-2.\u003c/li\u003e\n\u003cli\u003eZidere, V., et al., \u003cem\u003eComparison of echocardiographic findings in fetuses at less than 15 weeks\u0026apos; gestation with later cardiac evaluation.\u003c/em\u003e Ultrasound Obstet Gynecol, 2013. \u003cstrong\u003e42\u003c/strong\u003e(6): p. 679-86.\u003c/li\u003e\n\u003cli\u003eMacColl, C.E., et al., \u003cem\u003eFactors associated with in utero demise of fetuses that have underlying cardiac pathologies.\u003c/em\u003e Pediatr Cardiol, 2014. \u003cstrong\u003e35\u003c/strong\u003e(8): p. 1403-14.\u003c/li\u003e\n\u003cli\u003eJepson, B.M., et al., \u003cem\u003ePregnancy loss in major fetal congenital heart disease: incidence, risk factors and timing.\u003c/em\u003e Ultrasound Obstet Gynecol, 2023. \u003cstrong\u003e62\u003c/strong\u003e(1): p. 75-87.\u003c/li\u003e\n\u003cli\u003eRudolph, A.M., \u003cem\u003eCirculatory changes during gestational development of the sheep and human fetus.\u003c/em\u003e Pediatr Res, 2018. \u003cstrong\u003e84\u003c/strong\u003e(3): p. 348-351.\u003c/li\u003e\n\u003cli\u003eRudolph, A.M. and M.A. Heymann, \u003cem\u003eThe fetal circulation.\u003c/em\u003e Annu Rev Med, 1968. \u003cstrong\u003e19\u003c/strong\u003e: p. 195-206.\u003c/li\u003e\n\u003cli\u003eSun, L., et al., \u003cem\u003eUnderstanding Fetal Hemodynamics Using Cardiovascular Magnetic Resonance Imaging.\u003c/em\u003e Fetal Diagn Ther, 2020. \u003cstrong\u003e47\u003c/strong\u003e(5): p. 354-362.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTABLE 1:\u0026nbsp;\u003c/strong\u003eMaternal and fetal characteristics stratified by high acuity neonatal care\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003eHigh acuity neonatal care\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003eNo high acuity \u0026nbsp;neonatal care\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003eAll patients\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"8\" valign=\"top\"\u003e\n \u003cp\u003eMaternal Factors\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eMaternal age (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e35.1\u0026plusmn;5.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e33.6\u0026plusmn;4.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e34.4\u0026plusmn;5.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003ePrevious livebirths\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.8\u0026plusmn;1.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.9\u0026plusmn;1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e123\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.8\u0026plusmn;1.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eTwin pregnancy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (10.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e1 (1.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8 (6.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eMaternal diabetes (all types) \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e10 (14.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3 (4.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e134\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e13 (9.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAnti-Ro positive antibody\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e4 (5.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e27 (42.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e31 (23.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eFertility treatments \u0026nbsp;(IVF or IUI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e11 (19.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8 (14.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e134\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e19 (14.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"8\" valign=\"top\"\u003e\n \u003cp\u003eFetal Factors\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eIncreased NT (3.5 mm and above)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (22.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e4 (14.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e13 (19.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eGenetic Abnormalities\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e12 (34.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e19 (30.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eNon-cardiac abnormalities\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e15 (26.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (16.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e111\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e24 (21.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; Fetal growth restriction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e12 (20.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3 (4.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e122\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e15 (12.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; Placental abnormalities\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8 (14.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e1 (1.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e111\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (8.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.03\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eSeverity of structural CHD (major)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e39 (65.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e15 (41.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e54 (56.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.03\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eHoffman cardiac severity \u0026gt; mild\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e19 (35.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e18 (58.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e37 (43.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"8\" valign=\"top\"\u003e\n \u003cp\u003eCardiac Diagnoses\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eRight heart disease\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e29 (41.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e17 (26.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e46 (34.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eLeft heart disease\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (12.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (6.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.003\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAtrioventricular septal defect\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (10.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e1 (1.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8 (6.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eRight ventricle \u0026nbsp;hypoplasia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e17 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8 (12.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e25 (18.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAbnormal right ventricle function\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e9 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e1 (1.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e10 (7.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.02\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eLeft ventricle hypoplasia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e13 (18.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e4 (6.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e17 (12.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.04\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAbnormal tricuspid valve\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e14 (20.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (10.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e21 (15.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAbnormal pulmonary valve\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (10.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3 (4.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e134\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e10 (7.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAbnormal aortic valve\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e6 (8.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e1 (1.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7 (5.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eAbnormal mitral valve\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e4 (5.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e2 (3.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e134\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e6 (4.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003ePericardial Effusion\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e10 (14.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3 (4.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e13 (9.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"8\" valign=\"top\"\u003e\n \u003cp\u003ePostnatal Outcome\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eSex (male)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e28 (44.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e27 (47.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e120\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e55 (45.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eGestational age at birth (weeks)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e35.3\u0026plusmn;5.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e38.0\u0026plusmn;2.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e126\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e36.5\u0026plusmn;4.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eBirth weight (kg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e2.4\u0026plusmn;1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3.0\u0026plusmn;0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e114\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e2.7\u0026plusmn;0.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eApgar 1 min\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7.1\u0026plusmn;2.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8.6\u0026plusmn;1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e7.8\u0026plusmn;2.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eApgar 5 min\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8.2\u0026plusmn;1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8.9\u0026plusmn;0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e8.5\u0026plusmn;1.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"44.791666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eCardiac intervention (postnatal)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e27 (42.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"4.166666666666667%\" valign=\"top\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e3 (6.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.208333333333333%\" valign=\"top\"\u003e\n \u003cp\u003e109\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e30 (27.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.416666666666666%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eSLE, systemic lupus erythematosus; NT, nuchal thickness; GU= genitourinary; \u0026nbsp;; VSD, ventricular septal defect; SVD, \u0026nbsp;spontaneous vaginal delivery; IVF, in vitro fertilization; IUI, in utero insemination\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2:\u0026nbsp;\u003c/strong\u003eEvaluation of the prediction models for fetal or neonatal death, high acuity neonatal care and favourable outcome algorithms. PPV and NPV represent positive and negative predictive values, AUC represent the area under the receiver-operating characteristic curve.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"748\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.04127829560586%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutcome\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of patients\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutcome prevalence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of features in the model\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensitivity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSpecificity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.04127829560586%\"\u003e\n \u003cp\u003ePerinatal \u0026nbsp;death\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.04127829560586%\"\u003e\n \u003cp\u003eHigh acuity neonatal care\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.04127829560586%\"\u003e\n \u003cp\u003eFavourable outcome\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.119840213049267%\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"pediatric-cardiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pedc","sideBox":"Learn more about [Pediatric Cardiology](http://link.springer.com/journal/246)","snPcode":"246","submissionUrl":"https://submission.nature.com/new-submission/246/3","title":"Pediatric Cardiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Machine Learning, fetal cardiology, congenital heart disease, outcomes","lastPublishedDoi":"10.21203/rs.3.rs-4045996/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4045996/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBACKGROUND:\u003c/strong\u003e Prediction of outcomes following a prenatal diagnosis of congenital heart disease is challenging. Machine learning (ML) algorithms may be used to reduce clinical uncertainty and improve prognostic accuracy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMETHODS:\u003c/strong\u003e We performed a pilot study to train ML algorithms to predict postnatal outcomes based on clinical data. Specific objectives were to predict 1) in-utero or neonatal death, 2) high-acuity neonatal care and 3) favourable outcomes. We included all fetuses with cardiac disease at Sunnybrook Health Sciences Centre, Toronto, Canada, from 2012 – 2021. Prediction models were created using the XgBoost algorithm (tree-based) with 5-fold cross validation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRESULTS:\u003c/strong\u003e Among 211 cases of fetal cardiac disease, 61 were excluded (39 terminations, 21 lost to follow-up, 1 isolated arrhythmia), leaving a cohort of 150 fetuses. Fifteen (10%) demised (10 neonates) and 70 (52%) of live births required high acuity neonatal care. Of those with clinical follow-up, 57/82 (70%) had a favourable outcome. Prediction models for live birth, high acuity neonatal care and favourable outcome had AUCs of 0.75, 0.82 and 0.72, respectively. The most important predictors for death were the presence of non-cardiac or genetic abnormalities and more severe structural heart disease. High acuity of postnatal care was predicted by increased nuchal thickness, lower gestational age (GA) and birthweight and favourable outcome was predicted by normal fetal right ventricular function, no tricuspid valve abnormalities, and normal GA/weight at birth.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCONCLUSION\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrediction models using ML provide good discrimination of key prenatal and postnatal outcomes among fetuses with congenital heart disease.\u003c/p\u003e","manuscriptTitle":"Machine learning to predict outcomes of fetal cardiac disease: a pilot study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-13 16:41:33","doi":"10.21203/rs.3.rs-4045996/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-04-04T17:47:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-04-02T22:39:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"7f2a4656-34bb-47db-ab55-17f94229bb71","date":"2024-03-12T19:26:14+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-03-12T19:11:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-03-10T06:12:29+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-03-10T06:12:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"Pediatric Cardiology","date":"2024-03-08T14:57:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"pediatric-cardiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pedc","sideBox":"Learn more about [Pediatric Cardiology](http://link.springer.com/journal/246)","snPcode":"246","submissionUrl":"https://submission.nature.com/new-submission/246/3","title":"Pediatric Cardiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"89318ed7-497d-48b1-8262-61d0cf2e3f1b","owner":[],"postedDate":"March 13th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-05-14T21:21:46+00:00","versionOfRecord":{"articleIdentity":"rs-4045996","link":"https://doi.org/10.1007/s00246-024-03512-x","journal":{"identity":"pediatric-cardiology","isVorOnly":false,"title":"Pediatric Cardiology"},"publishedOn":"2024-05-09 21:17:48","publishedOnDateReadable":"May 9th, 2024"},"versionCreatedAt":"2024-03-13 16:41:33","video":"","vorDoi":"10.1007/s00246-024-03512-x","vorDoiUrl":"https://doi.org/10.1007/s00246-024-03512-x","workflowStages":[]},"version":"v1","identity":"rs-4045996","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4045996","identity":"rs-4045996","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}