Early predction of preeclampsia risk in pregnant women unilizing random forest and particle swarm optimization techniques

R. Topan Aditya Rahman

QR Code Link :
Type :	Thesis
Subject :	Q Science
Main Author :	R. Topan Aditya Rahman
Title :	Early predction of preeclampsia risk in pregnant women unilizing random forest and particle swarm optimization techniques
Hits :	2

Place of Production :	Tanjong Malim
Publisher :	Fakulti Seni, Komputeran dan Industri Kreatif
Year of Publication :	2024
Corporate Name :	Perpustakaan Tuanku Bainun
PDF Guest :	Click to view PDF file
PDF Full Text :	Click to view PDF file

Abstract : Perpustakaan Tuanku Bainun

The main purpose of this study is to predict early preeclampsia using machine learning algorithms with Particle Swarm Optimization. Preeclampsia is one of the causes of maternal mortality, in the last two decades, there has been no significant decrease in the incidence of preeclampsia. The magnitude of this problem has an impact on the mother during pregnancy and childbirth. The entire population in this study was pregnant women. The entire population in this study was pregnant women. The samples in this study were 504 pregnant women from the medical records of Ansari Saleh General Hospital Banjarmasin in 2022. The algorithm used in this study using eXtreme Gradient Boosting, Adaptive Boosting, Random Forest, Logistic Regression, and for optimization algorithm using Particle Swarm Optimization. Based on the result, Random Forest was the best model with an accuracy rate of 96.08%. The variables that most influence the incidence of preeclampsia are the history of preeclampsia, a history of hypertension, a history of caesarean section delivery, and a history of diabetes mellitus. This successful evaluation of model development provides implications to help health workers in carrying out pregnancy screening, the trust of expectants in the service quality given surely affects society to utilize technology-based service rather than the conventional one. It recommends developing a prototype application for the early detection of preeclampsia using machine learning technology to assist healthcare professionals in delivering optimal antenatal care and transitioning to technology-based pregnancy monitoring as a proactive measure for pregnant women, ultimately helping to prevent preeclampsia.

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.

Back to search page