Using Innovative Machine Learning Methods to Screen and Identify Predictors of Congenital Heart Diseases

• Type of Material: Article
• Authors: Yanji Qu, Guangdong Academy of Medical Sciences; Xinlei Deng, State University of New York Albany; Shao Lin, State University of New York Albany; Fengzhen Han, Guangdong Academy of Medical Sciences; Howard Chang, Emory University; Yanqiu Ou, Guangdong Academy of Medical Sciences; Zhiqiang Nie, Guangdong Academy of Medical Sciences; Jinzhuang Mai, Guangdong Academy of Medical Sciences; Ximeng Wang, Guangdong Academy of Medical Sciences; Xiangmin Gao, Guangdong Academy of Medical Sciences; Yong Wu, Guangdong Academy of Medical Sciences; Jimei Chen, Guangdong Academy of Medical Sciences; Jian Zhuang, Guangdong Academy of Medical Sciences; Ian Ryan, State University of New York Albany; Xiaoqing Liu, Guangdong Academy of Medical Sciences
• Language: English
• Date: 2022-01-07
• Publisher: FRONTIERS MEDIA SA
• Publication Version: Final Published Version
• Copyright Statement: © 2022 Qu, Deng, Lin, Han, Chang, Ou, Nie, Mai, Wang, Gao, Wu, Chen, Zhuang, Ryan and Liu.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.3389/fcvm.2021.797002
• Title of Journal or Parent Work: FRONTIERS IN CARDIOVASCULAR MEDICINE
• Volume: 8
• Start Page: 797002
• End Page: 797002
• Grant/Funding Information: This study was supported by grants from the Science and Technology Planning Project of Guangdong Province, China (Nos. 2019B020230003, 2017A070701013, 2017B090904034, and 2017030314109), National Key Research and Development Program (No. 2018YFC1002600), Guangdong Peak Project (No. DFJH201802), Guangdong Provincial Key Laboratory of South China Structural Heart Disease (No. 2012A061400008), National Natural Science Foundation of China (No. 81903287), and Natural Science Foundation of Guangdong Province (Nos. 2018A030313785 and 2018A030313329).
• Keywords: CLASSIFICATION; RISK; CURRENT KNOWLEDGE; prediction; congenital heart disease; machine learning; ASSOCIATION; clinical indicators; DEFECTS; Life Sciences & Biomedicine; SCIENTIFIC STATEMENT; Cardiac & Cardiovascular Systems; laboratory tests; Cardiovascular System & Cardiology; PREGNANCY; HYPERGLYCEMIA; Science & Technology; CARDIOVASCULAR-DISEASE; URIC-ACID
• Research Categories: Environmental Sciences; Biology, Biostatistics

Abstract

Objective: Congenital heart diseases (CHDs) are associated with an extremely heavy global disease burden as the most common category of birth defects. Genetic and environmental factors have been identified as risk factors of CHDs previously. However, high volume clinical indicators have never been considered when predicting CHDs. This study aimed to predict the occurrence of CHDs by considering thousands of variables from self-reported questionnaires and routinely collected clinical laboratory data using machine learning algorithms. Methods: We conducted a birth cohort study at one of the largest cardiac centers in China from 2011 to 2017. All fetuses were screened for CHDs using ultrasound and cases were confirmed by at least two pediatric cardiologists using echocardiogram. A total of 1,127 potential predictors were included to predict CHDs. We used the Explainable Boosting Machine (EBM) for prediction and evaluated the model performance using area under the Receive Operating Characteristics (ROC) curves (AUC). The top predictors were selected according to their contributions and predictive values. Thresholds were calculated for the most significant predictors. Results: Overall, 5,390 mother-child pairs were recruited. Our prediction model achieved an AUC of 76% (69-83%) from out-of-sample predictions. Among the top 35 predictors of CHDs we identified, 34 were from clinical laboratory tests and only one was from the questionnaire (abortion history). Total accuracy, sensitivity, and specificity were 0.65, 0.74, and 0.65, respectively. Maternal serum uric acid (UA), glucose, and coagulation levels were the most consistent and significant predictors of CHDs. According to the thresholds of the predictors identified in our study, which did not reach the current clinical diagnosis criteria, elevated UA (>4.38 mg/dl), shortened activated partial thromboplastin time (<33.33 s), and elevated glucose levels were the most important predictors and were associated with ranges of 1.17-1.54 relative risks of CHDs. We have developed an online predictive tool for CHDs based on our findings that may help screening and prevention of CHDs. Conclusions: Maternal UA, glucose, and coagulation levels were the most consistent and significant predictors of CHDs. Thresholds below the current clinical definition of "abnormal" for these predictors could be used to help develop CHD screening and prevention strategies.