An Appropriate Feature Selection Technique for Use on Socio-Demographic Predictor Variables to Enable Early Detection of Preeclampsia: A Review of Literature

Abstract

Preeclampsia is categorized by the World Health Organization as one of the leading causes of high morbidity and mortality in infant and mothers around the world. It accounts for between 3% to 5% of all pregnancy related complications reported worldwide. This condition is much higher among women aged between 30 and 40 years in developing nations especially those in the sub-Saharan region, where the figures range between 5.6% to 6.5% of all reported pregnancies. Preeclampsia is a condition normally detected in the third trimester of pregnancy that is characterized by high risk factors such as sudden High Blood Pressure, High levels of protein in Urine, Chronic kidney disease and Type 1 or 2 diabetes. If preeclampsia is not detected early, it can advance to eclampsia or result to maternal and fetal death. This study sought to identify the optimal features as predictors to enable early detection of preeclampsia through a systematic review of relevant literature. The predictors under consideration were; Maternal age, Occupation, Education, ANC Attendance, BMI, Blood Pressure, Medical History, Urine dipstick, Gravida, Ethnicity, Gestation weeks as identified from literature.

Keywords: Ante natal care service, Preeclampsia, feature engineering, socio-demographic features, machine learning

DOI: 10.7176/CEIS/13-4-02

Publication date:August 31^st 2022