AI-DRIVEN PREDICTIVE MODELS FOR ACCURATE LBW CASE PREDICTIONS
Main Article Content
Abstract
The study focuses on using machine learning to identify newborns at risk of low birth weight (LBW), a key indicator of neonatal illness and a predictor of future health issues. It highlights the significant link between a mother's health during pregnancy and her baby's birth weight. The research transforms this forecasting challenge into a binary classification task, distinguishing between LBW and non-LBW cases using supervised machine learning techniques. The model, which demonstrates enhanced accuracy, leverages health data from India. This data forms the basis for developing decision-making rules that can be applied in the context of predictive healthcare in smart cities. Additionally, a specialized screening tool, designed using this decision model, aims to support professionals in Obstetrics and Gynecology (OBG). The study emphasizes the importance of smart health informatics, predictive analytics, and machine learning in advancing maternal and neonatal healthcare.