NutriScreener: AI to scale early malnutrition screening
NutriScreener is a new AI method to help screen for child malnutrition quickly and at scale—using ordinary photos captured from multiple poses.
How it works: it combines CLIP-based visual features, a retrieval “knowledge base” that surfaces context from similar cases, and a multi-pose graph attention network that reasons across views. This boosts robustness across populations and helps with class imbalance.
- Clinician ratings: 4.3/5 for accuracy and 4.6/5 for efficiency, suggesting suitability for low-resource settings.
- Performance: 0.79 recall, 0.82 AUC, and significantly lower errors when estimating anthropometric measures in unconstrained pediatric environments.
- Generalization: trained/tested on 2,141 children (AnthroVision) and evaluated on ARAN plus a new CampusPose dataset. Using demographically matched knowledge bases yielded up to +25% recall and up to 3.5 cm RMSE reduction across datasets.
Why it matters: earlier, scalable screening can enable faster intervention where tools and skilled staff are scarce.
Paper: https://arxiv.org/abs/2511.16566v1
Paper: https://arxiv.org/abs/2511.16566v1
Register: https://www.AiFeta.com
AI Healthcare GlobalHealth ComputerVision Malnutrition LowResource PublicHealth MachineLearning
