Related Experiment Video
Updated: May 8, 2026
Milk Collection in the Rat Using Capillary Tubes and Estimation of Milk Fat Content by Creamatocrit
Published on: December 16, 2015
The Application of Machine Learning in the Analysis of Macro and Micro-Nutrients of Human Milk: a Scoping Review
Ameneh Arzheh1, Memnon Seven1, Carrie-Ellen Briere2
1Elaine Marieb College of Nursing (EMCON), University of Massachusetts Amherst, Amherst, MA, United States.
Background:
Artificial intelligence, particularly machine learning, has significantly advanced the analysis of complex biological systems, including human milk-a vital source of nutrition for infant growth and development. Despite advancements, the application of machine learning for detailed nutritional profiling of human milk remains limited.
Objective:
This review systematically examines and synthesizes the existing literature on the application of machine learning to analyze and predict the macro- and micronutrient content of human milk.
Methods:
Following the Joanna Briggs Institute Methods Manual, a systematic search was conducted in November 2024 across 4 sources: MEDLINE (via PubMed), CINAHL, Scopus, and the ACM Digital Library. Eligible studies applied machine learning techniques to analyze and predict the composition of human milk.
Results:
Five studies published between 2021 and 2024 were included, employing cross-sectional (n = 4) and cohort (n = 1) designs. Sample sizes ranged from 6 to >1000 breast milk samples, collected through pumping, hand expression, or milk bank donations. A variety of machine learning algorithms were applied to predict and analyze human milk components. Models demonstrated strong predictive accuracy for macronutrients-particularly fat and protein-which consistently achieved low error rates and high R2 values. Predictions for lactose and energy were also promising, though somewhat less consistent across methods. Only one study addressed micronutrient prediction, yielding moderate results and highlighting a critical gap in the literature.
Conclusions:
Machine learning holds strong potential for advancing human milk research by improving nutrient composition analysis and uncovering associations with maternal and environmental factors. However, existing studies remain limited, particularly in the areas of micronutrient analysis and laboratory-based nutrient profiling. Future research should focus on larger and more diverse datasets, and adopt standardized methods to enhance reproducibility and unlock the full potential of machine learning for personalized infant nutrition and health.
