Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

nafis2508/maternal-neonatal-outcome-prediction

Domain:

healthcare

Record type:

dataset
Creator:
naf
Host:
Predicting Maternal and Neonatal Birth Outcomes using Machine Learning on 61,018 Healthcare Records from Kenya and Uganda # Maternal & Neonatal Outcome Prediction using Machine Learning ## Overview This project applies machine learning techniques to predict maternal and neonatal birth outcomes using healthcare facility delivery records collected from Kenya and Uganda. The objective is to identify factors associated with adverse pregnancy outcomes and evaluate the effectiveness of machine learning models in predicting birth outcomes. The project combines healthcare analytics, exploratory data analysis, feature engineering, class imbalance handling, and predictive modeling to support data-driven healthcare decision-making. --- ## Key Project Metrics | Metric | Value | |----------|----------| | Records Analysed | 61,018 | | Countries | Kenya & Uganda | | Features Used | 13 | | Machine Learning Models | 4 | | Best Model | Random Forest | | Best AUROC | 0.905 | --- ## Business Problem Maternal and neonatal mortality remain major public health challenges in many low-resource regions. Healthcare providers require reliable methods to identify pregnancies at higher risk of adverse outcomes so that preventative interventions can be implemented earlier. This project investigates whether machine learning models can accurately predict birth outcomes using maternal characteristics, gestational information, and neonatal indicators. --- ## Dataset Information ### Source Healthcare facility delivery records collected from: - Kenya - Uganda ### Dataset Size - 61,018 delivery records - Originally 22 attributes - Reduced to 13 key predictive features after preprocessing ### Key Variables | Feature | Description | |----------|------------| | Country | Delivery country | | Referral Status | Hospital referral information | | Sex | Baby sex | | Multiple Birth | Multiple delivery indicator | | Abortion History | Documented abortion indicator | | IUFD | Intrauterine fetal demise indicator | | Mother's Age Category | Maternal age groups | | Gestational Age Category | Pregnancy duration catego …

Visit

github.com

Tags

data-sciencedecision-treeedahealthcare-aihealthcare-analyticsmachine-learningmaternal-healthpredictive-modelingpythonrandom-forest+1

Similar

Comparison of neonatal outcome.Lumumba1992/Neonatal-Mortality-PredictionKnowledge-guided Transfer Prediction In Underrepresented Populations: A GRU-D-Static Framework For Maternal And Neonatal OutcomesToniaNiles/maternal-risk-prediction-refactoredoogunjobi7973-byte/maternal-mortality-predictionMomahmoses/nigeria-maternal-mortality-prediction

Comparison of neonatal outcome.

Background

Women of reproductive age constitute a significant proportion of the global

Lumumba1992/Neonatal-Mortality-Prediction

An Explainable AI Framework for Neonatal Mortality Risk Prediction in Kenya: Enhancing Clinical Deci

Knowledge-guided Transfer Prediction In Underrepresented Populations: A GRU-D-Static Framework For Maternal And Neonatal Outcomes

Integrating summary-level scientific knowledge into neural network models provides a practical strat

ToniaNiles/maternal-risk-prediction-refactored

Refactored MSc dissertation project exploring maternal health vulnerability in Nigeria using machine

oogunjobi7973-byte/maternal-mortality-prediction

Early Maternal Mortality Prediction System for Rural Healthcare Systems in Nigeria # Early Maternal

Momahmoses/nigeria-maternal-mortality-prediction

||XGBoost and SHAP maternal mortality risk scoring for Nigerian PHCs with a DHIS2-compatible REST AP