An Automated System for Early Prediction of Miscarriage in the First Trimester Using Machine Learning  

作　　者：Sumayh S.Aljameel Malak Aljabri Nida Aslam Dorieh M.Alomari Arwa Alyahya Shaykhah Alfaris Maha Balharith Hiessa Abahussain Dana Boujlea Eman S.Alsulmi 

机构地区：[1]Department of Computer Science,College of Computer Science and Information Technology,Imam Abdulrahman Bin Faisal University,P.O.Box 1982,Dammam,31441,Saudi Arabia [2]Computer Science Department,College of Computer and Information Systems,Umm Al-Qura University,Makkah,21955,Saudi Arabia [3]Department of Computer Engineering,College of Computer Science and Information Technology,Imam Abdulrahman Bin Faisal University,P.O.Box 1982,Dammam,31441,Saudi Arabia [4]Department of Obstetrics and Gynecology,College of Medicine,Imam Abdulrahman Bin Faisal University,Dammam,Saudi Arabia

出　　处：《Computers, Materials & Continua》2023年第4期1291-1304,共14页计算机、材料和连续体（英文）

摘　　要：Currently, the risk factors of pregnancy loss are increasing andare considered a major challenge because they vary between cases. The earlyprediction of miscarriage can help pregnant ladies to take the needed careand avoid any danger. Therefore, an intelligent automated solution must bedeveloped to predict the risk factors for pregnancy loss at an early stage toassist with accurate and effective diagnosis. Machine learning (ML)-baseddecision support systems are increasingly used in the healthcare sector andhave achieved notable performance and objectiveness in disease predictionand prognosis. Thus, we developed a model to help obstetricians predictthe probability of miscarriage using ML. And support their decisions andexpectations about pregnancy status by providing an easy, automated way topredict miscarriage at early stages using ML tools and techniques. Althoughmany published papers proposed similar models, none of them used Saudiclinical data. Our proposed solution used ML classification algorithms tobuild a miscarriage prediction model. Four classifiers were used in this study:decision tree (DT), random forest (RF), k-nearest neighbor (KNN), andgradient boosting (GB). Accuracy, Precision, Recall, F1-score, and receiveroperating characteristic area under the curve (ROC-AUC) were used to evaluatethe proposed model. The results showed that GB overperformed the otherclassifiers with an accuracy of 93.4% and ROC-AUC of 97%. This proposedmodel can assist in the early identification of at-risk pregnant women to avoidmiscarriage in the first trimester and will improve the healthcare sector inSaudi Arabia.

关 键 词：MISCARRIAGE PREGNANCY ABORTION machine learning gradient boosting 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]