AI-Based Expanded BMI System for Early Malnutrition Risk Detection

This paper has presented an automated, image-based malnutrition detection system applying a fine-tuned YOLOv5m deep learning model to classify children as malnourished or healthy from a single photograph. The system integrates detection, classification, and nutritional recommendation into a unified web application deployable on standard consumer hardware without specialist involvement. Experimental evaluation demonstrated 94.7% accuracy, 93.2% precision, 95.1% recall, and an AUC-ROC of 0.974, surpassing ResNet-50 and VGG-16 baselines by statistically meaningful margins. The system runs at 187 ms per image on an Intel Core i5 with 1 GB RAM, confirming practical feasibility for resource-limited primary health settings.

[1] A. R. Lakshminarayanan et al., “Malnutrition detection using convolutional neural network,” in Proc. 7th Int. Conf. Bio Signals, Images, and Instrumentation (ICBSII), IEEE, 2021, pp. 1–5. [2] S. Ankalaki, V. G. Biradar, K. K. Naik, and G. S. Hukkeri, “A deep learning approach for malnutrition detection,” Int. J. Online Biomed. Eng. (iJOE), vol. 20, no. 6, pp. 116–138, 2024. [3] V. G. Biradar and K. K. Naik, “Detection of malnutrition in children using deep learning model,” in Smart Trends in Computing and Communications (SmartCom 2024), vol. 946, Springer, 2024. [4] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, real-time object detection,” in Proc. IEEE CVPR, Las Vegas, NV, Jun. 2016, pp. 779–788. [5] G. Jocher et al., “Ultralytics YOLOv5,” Zenodo, 2021, doi: 10.5281/zenodo.3908559. [6] T. Diwan, G. Anirudh, and J. V. Tembhurne, “Object detection using YOLO: challenges, architectural successors, datasets and applications,” Multimed. Tools Appl., vol. 82, no. 6, pp. 9243–9275, 2023. [7] R. Y. Toledo, A. A. Alzahrani, and L. Martinez, “A food recommendation system considering nutritional information and user preferences,” IEEE Access, vol. 7, pp. 96695–96711, 2019. [8] M. Mansouri et al., “Deep learning for food image recognition and nutrition analysis,” SN Comput. Sci., vol. 4, no. 513, 2023.