The triple burden of malnutrition and the role of machine learning: A systematic review

BackgroundMalnutrition remains a key global health concern encompassing the triple burden of undernutrition (stunting and wasting), micronutrient deficiencies, and overnutrition. These situations persist across all life phases and regions, significantly contributing to diseases, mortality, and impaired development. With increasing data availability, machine learning has emerged as a powerful tool capable of identifying complex, and diverse determinants of malnutrition and providing early, data-driven predictions for interventions.ObjectiveThe primary objective of this article is to provide a systematically review of existing literature to identify the determinants and contributing factors of malnutrition across different age groups and to highlight the application of machine learning algorithms in predicting malnutrition.MethodsA systematic literature review was conducted using multiple databases. The review analysed determinants of malnutrition and assessed the use of various ML algorithms such as decision trees, random forests, support vector machines, and neural networks in predicting malnutrition. Studies included were peer-reviewed and focused on assessable features such as anthropometric, socioeconomic, dietary, and environmental factors.ResultsThe review highlights that the triple burden of malnutrition continues as a global health challenge. Socioeconomic status, demographic characteristics, dietary patterns, maternal education, and health indicators were identified as chief determinants. ML models have demonstrated high predictive accuracy in identifying at-risk populations by studying multidimensional relationships that traditional methods often overlook. Among these, decision trees, random forests, and neural networks performed constantly well across diverse datasets and contexts.ConclusionThis review founds that machine learning plays a transformative role in addressing malnutrition by enabling early prediction, identification of risk factors, and policymaking. However, current studies are largely centred on children and pregnant women, with limited focus on adolescents especially adolescent girls who are equally vulnerable to undernutrition, micronutrient deficiencies, and obesity. Future research should expand to include adolescent populations and employ real-time, primary data from wearable devices, mobile health applications, and IoT-based nutrition monitoring tools. The adoption of explainable AI and deep learning will further enhance model interpretability and scalability. Collaborating these advanced technologies will reinforce global efforts to achieve sustainable nutrition and health equity.