Vol. 7, Issue 1, Part A (2025)

Background and Objectives: Low birth weight (LBW) is a critical indicator of maternal health and prenatal care quality worldwide. This study aimed to develop a predictive model using artificial intelligence (AI) to identify risk factors contributing to LBW.
Methods: A case-control study design was adopted, comparing 100 postnatal mothers with LBW neonates (cases) to 200 postnatal mothers with normal-weight neonates (controls). Logistic regression, enhanced by AI, was employed to develop the predictive model.
Results: The logistic regression model identified significant risk factors for LBW, including inadequate weight gain during pregnancy (<9 kg, p< 0.001), fetal complications during pregnancy (p< 0.001), maternal height <145 cm (p = 0.001), gestational age <37 weeks (p = 0.034), multiple pregnancies (p = 0.044), and maternal weight <45 kg (p = 0.048). The model demonstrated a high accuracy of 90%, with an AUC of 0.91, indicating excellent discriminatory power to distinguish between LBW and normal-weight neonates.
Conclusion: Most identified risk factors are modifiable through regular prenatal care and targeted interventions. Integrating this AI-driven predictive model into hospital information systems and public health programs enables early identification and proactive management of risk factors, significantly reducing LBW incidence and improving neonatal outcomes.