The Food, Nutritional Biochemistry and Health research group of the Universidad Europ0ea del Atlántico (European Atlantic University, UNEATLANTICO) is leading a publication in a new article in the prestigious journal Nutrients (Q1).
The article entitled «Image-Based Dietary Energy and Macronutrients Estimation with ChatGPT-5: Cross-Source Evaluation Across Escalating Context Scenarios» aims to study whether or not the ChatGPT-5 version has the ability to estimate calories and (proteins, carbohydrates and fats) from food dishes using only photos, or if their analysis improves when context is added to the photo. The context could be descriptions of the food or a list of ingredients.
The article was written by Marcela Rodríguez, graduate student in Human Nutrition and Dietetics (NHD) from UNEATLANTICO, in collaboration with Gustavo Martín del Campo Becerra, Sandra Sumalla, Jorge Crespo and Iñaki Elio.
The article comes from Rodríguez’s Final Grade Work. Due to the quality of the work, the research team decided to expand it into a scientific article suitable for international dissemination, thus reinforcing the commitment of the university to promote applied research from the early stages of training.
For the development of the study, 195 dishes were used and the results showed that ChatGPT-5 generates a better analysis based on the amount of information provided, image quality, source variability and complexity of the dish. However, the conclusion was that ChatGPT-5 can give a good analysis of the calories and macronutrients contained in a dish.
The article can be consulted via the following link: https://doi.org/10.3390/nu17223613
To read more about research articles from our community, you can consult the UNEATLANTICO Repository or the research news portal.