UPSI Digital Repository (UDRep)
|
UPSI Digital Repository (UDRep)
|
|
|
|
||||||||||||||||||||||||||
| Abstract : Universiti Pendidikan Sultan Idris |
| Premature birth remains a significant challenge in maternal health services, underscoring the need for effective predictive models to enable early detection and intervention. This study examined the performance of four machine learning algorithms: Decision Trees, Naive Bayes, KNN, and Random Forest in predicting preterm birth during pregnancy. Utilizing data collected from pregnant individuals, encompassing maternal health indicators and fetal development metrics, our model aimed to forecast the likelihood of preterm birth. We assessed the predictive ability of each model by evaluating metrics such as accuracy, precision, recall, sensitivity, specificity, and area under the curve (AUC). The results revealed variability in model performance, with Logistic Random Forest exhibiting strong performance. This suggests its potential utility in clinical settings for the early detection and intervention of preterm pregnancy. Our study contributes to advancements in predictive modelling within maternal health services, aiming to enhance maternal and fetal health outcomes through the early identification of preterm birth. © 2024, Collegium Basilea. All rights reserved. |
| This material may be protected under Copyright Act which governs the making of photocopies or reproductions of copyrighted materials. You may use the digitized material for private study, scholarship, or research. |