Eat4U: Calorie Management Based on Predictive Machine Learning Models and Instance Segmentation for University Students in Metropolitan Lima

Ramirez S.; Quispe J.; Navarrete E. · 2025 3rd Cognitive Models and Artificial Intelligence Conference, AICCONF 2025 - Proceedings (2025) · DOI: 10.1109/aicconf64766.2025.11064276

Resumen

Caloric imbalance among university students has emerged as a critical public health concern in the modern era, with significant implications for their long-term health and well-being. The lack of nutritional knowledge and awareness among young people often contributes to this problem. Although previous research has created nutrition applications to help people understand the nutritional value of foods, there is a notable lack of applications expressly designed to help university students in metropolitan Lima manage their caloric intake. To address this gap, we propose Eat4U, an intelligent mobile application aimed at providing students with essential information about the nutritional content, the base model used is Mask R-CNN ResNet50 FPN V2 and the results obtained through finetuning, and incremental learning resulted in a Mean-IoU of 86%, thereby increasing their awareness and understanding of nutrition. The main feature of the application is the use of the cell phone camera and machine learning models equipped with the PyTorch library, allowing users to scan meals and display detailed information on nutrients and calories to manage their caloric intake. Additionally, a personalized recommendation system is offered based on each user's tastes, goals, and objectives, helping them choose healthier food options. The application also includes an interactive log and a calendar of eating habits to monitor progress and encourage the adoption of healthy lifestyles.

Instituciones: Universidad Peruana de Ciencias Aplicadas