Peer-reviewed article
Machine Learning Techniques to Predict Fetal Nutritional Status.
Mduma N, Laizer H ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This study developed a machine learning model to predict fetal nutritional status using maternal and clinical data from 11,703 pregnant women in Tanzania, where malnutrition affects over one-third of children under 5. The researchers compared five algorithms and found that LightGBM performed best, identifying key predictors such as maternal age, weight, and hemoglobin level. The model was deployed in a mobile application that achieved 93% user satisfaction among health workers and parents. The tool aims to enable early identification of at-risk pregnancies and support timely interventions to reduce child malnutrition.
Key points
- LightGBM algorithm achieved the best predictive performance for fetal nutritional status
- Key predictors were maternal age, weight, fetal age, hemoglobin level, meals per day, medical history, and education level
- Model was trained on data from 11,703 pregnant women in Tanzania
- Mobile application deployment achieved 93% user satisfaction among respondents
- Tool designed to enable early risk identification and timely intervention in low-resource settings
Limitations
The document does not report the actual predictive accuracy metrics (sensitivity, specificity, or AUC scores) of the LightGBM model, making it impossible to assess how well it performs in practice. There is no information on whether the predicted early interventions actually improved maternal and fetal outcomes or reduced childhood malnutrition.
Abstract
Objectives Malnutrition remains the leading cause of child mortality in Tanzania, with over 34% of children under 5 years of age affected by stunting and approximately 5% experiencing acute malnutrition. This study aimed to develop a machine learning model to predict fetal nutritional status using maternal and clinical data, thereby enabling early risk identification for health workers and parents and facilitating timely intervention. To enhance practical applicability, the model was deployed within a mobile application to provide accessible, real-time predictions that support prompt clinical and behavioral responses. Methods Using a dataset collected in Tanzania, the performance of multiple binary classification algorithms-logistic regression, multi-layer perceptron, random forest, extreme gradient boosting, and light gradient boosting machine (LightGBM)-was compared using the geometric mean and F-measure. These models were trained on clinical data from 11,703 pregnant women to predict fetal nutritional status based on maternal and clinical variables. Results The results indicated that the LightGBM algorithm achieved the best overall performance in predicting fetal nutritional status. The most influential predictors included maternal age, weight, fetal age, hemoglobin level, number of meals per day, medical history, and education level. Additionally, 93% of respondents reported satisfaction with the application's predictive functionality, supporting its potential utility for early intervention in low-resource settings. Conclusions These findings highlight the potential of data-driven approaches to address public health challenges in maternal and child health. The proposed model may enable healthcare providers to make timely, informed decisions that improve maternal and fetal outcomes, ultimately contributing to the reduction of child malnutrition in Tanzania.
Findings the Hub has extracted
Each claim below is labelled by its type of design and links back to this report.
- Intervention
- Machine learning model (LightGBM) deployed in a mobile application to predict fetal nutritional status based on maternal and clinical variables
- Population
- Pregnant women in Tanzania (training data from 11,703 women)
Key finding
LightGBM algorithm achieved best performance compared to logistic regression, multi-layer perceptron, random forest, and extreme gradient boosting
N et al. · Europe PMC · 31 Jul 2026