Mestrado em Direcção Estratégica em Tecnologias de Informação

<a href="/26722/1/nutrients-18-00257.pdf" class="ep_document_link"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Innovative Application of Chatbots in Clinical Nutrition Education: The E+DIEting_Lab Experience in University Students

Background/Objectives: The growing integration of Artificial Intelligence (AI) and chatbots in health professional education offers innovative methods to enhance learning and clinical preparedness. This study aimed to evaluate the educational impact and perceptions in university students of Human Nutrition and Dietetics, regarding the utility, usability, and design of the E+DIEting_Lab chatbot platform when implemented in clinical nutrition training. Methods: The platform was piloted from December 2023 to April 2025 involving 475 students from multiple European universities. While all 475 students completed the initial survey, 305 finished the follow-up evaluation, representing a 36% attrition rate. Participants completed surveys before and after interacting with the chatbots, assessing prior experience, knowledge, skills, and attitudes. Data were analyzed using descriptive statistics and independent samples t-tests to compare pre- and post-intervention perceptions. Results: A total of 475 university students completed the initial survey and 305 the final evaluation. Most university students were females (75.4%), with representation from six languages and diverse institutions. Students reported clear perceived learning gains: 79.7% reported updated practical skills in clinical dietetics and communication were updated, 90% felt that new digital tools improved classroom practice, and 73.9% reported enhanced interpersonal skills. Self-rated competence in using chatbots as learning tools increased significantly, with mean knowledge scores rising from 2.32 to 2.66 and skills from 2.39 to 2.79 on a 0–5 Likert scale (p < 0.001 for both). Perceived effectiveness and usefulness of chatbots as self-learning tools remained positive but showed a small decline after use (effectiveness from 3.63 to 3.42; usefulness from 3.63 to 3.45), suggesting that hands-on experience refined, but did not diminish, students’ overall favorable views of the platform. Conclusions: The implementation and pilot evaluation of the E+DIEting_Lab self-learning virtual patient chatbot platform demonstrate that structured digital simulation tools can significantly improve perceived clinical nutrition competences. These findings support chatbot adoption in dietetics curricula and inform future digital education innovations.

Producción Científica

Iñaki Elío Pascual mail inaki.elio@uneatlantico.es, Kilian Tutusaus mail kilian.tutusaus@uneatlantico.es, Imanol Eguren García mail imanol.eguren@uneatlantico.es, Álvaro Lasarte García mail , Arturo Ortega-Mansilla mail arturo.ortega@uneatlantico.es, Thomas Prola mail thomas.prola@uneatlantico.es, Sandra Sumalla Cano mail sandra.sumalla@uneatlantico.es,

Elío Pascual

<a class="ep_document_link" href="/17880/1/nutrients-17-03613.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Image-Based Dietary Energy and Macronutrients Estimation with ChatGPT-5: Cross-Source Evaluation Across Escalating Context Scenarios

Background/Objectives: Estimating energy and macronutrients from food images is clinically relevant yet challenging, and rigorous evaluation requires transparent accuracy metrics with uncertainty and clear acknowledgement of reference data limitations across heterogeneous sources. This study assessed ChatGPT-5, a general-purpose vision-language model, across four scenarios differing in the amount and type of contextual information provided, using a composite dataset to quantify accuracy for calories and macronutrients. Methods: A total of 195 dishes were evaluated, sourced from Allrecipes.com, the SNAPMe dataset, and Home-prepared, weighed meals. Each dish was evaluated under Case 1 (image only), Case 2 (image plus standardized non-visual descriptors), Case 3 (image plus ingredient lists with amounts), and Case 4 (replicates Case 3 but excluding the image). The primary endpoint was kcal Mean Absolute Error (MAE); secondary endpoints included Median Absolute Error (MedAE) and Root Mean Square Error (RMSE) for kcal and macronutrients (protein, carbohydrates, and lipids), all reported with 95% Confidence Intervals (CIs) via dish-level bootstrap resampling and accompanied by absolute differences (Δ) between scenarios. Inference settings were standardized to support reproducibility and variance estimation. Source stratified analyses and quartile summaries were conducted to examine heterogeneity by curation level and nutrient ranges, with additional robustness checks for error complexity relationships. Results and Discussion: Accuracy improved from Case 1 to Case 2 and further in Case 3 for energy and all macronutrients when summarized by MAE, MedAE, and RMSE with 95% CIs, with absolute reductions (Δ) indicating material gains as contextual information increased. In contrast to Case 3, estimation accuracy declined in Case 4, underscoring the contribution of visual cues. Gains were largest in the Home-prepared dietitian-weighed subset and smaller yet consistent for Allrecipes.com and SNAPMe, reflecting differences in reference curation and measurement fidelity across sources. Scenario-level trends were concordant across sources, and stratified and quartile analyses showed coherent patterns of decreasing absolute errors with the provision of structured non-visual information and detailed ingredient data. Conclusions: ChatGPT-5 can deliver practically useful calorie and macronutrient estimates from food images, particularly when augmented with standardized nonvisual descriptors and detailed ingredients, as evidenced by reductions in MAE, MedAE, and RMSE with 95% CIs across scenarios. The decline in accuracy observed when the image was omitted, despite providing detailed ingredient information, indicates that visual cues contribute meaningfully to estimation performance and that improvements are not solely attributable to arithmetic from ingredient lists. Finally, to promote generalizability, it is recommended that future studies include repeated evaluations across diverse datasets, ensure public availability of prompts and outputs, and incorporate systematic comparisons with non-artificial-intelligence baselines.

Producción Científica

Marcela Rodríguez- Jiménez mail , Gustavo Daniel Martín-del-Campo-Becerra mail , Sandra Sumalla Cano mail sandra.sumalla@uneatlantico.es, Jorge Crespo-Álvarez mail jorge.crespo@uneatlantico.es, Iñaki Elío Pascual mail inaki.elio@uneatlantico.es,

Rodríguez- Jiménez

<a href="/17885/1/s41598-025-26052-7.pdf" class="ep_document_link"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Dual-modality fusion for mango disease classification using dynamic attention based ensemble of leaf & fruit images

Mango is one of the most beloved fruits and plays an indispensable role in the agricultural economies of many tropical countries like Pakistan, India, and other Southeast Asian countries. Similar to other fruits, mango cultivation is also threatened by various diseases, including Anthracnose and Red Rust. Although farmers try to mitigate such situations on time, early and accurate detection of mango diseases remains challenging due to multiple factors, such as limited understanding of disease diversity, similarity in symptoms, and frequent misclassification. To avoid such instances, this study proposes a multimodal deep learning framework that leverages both leaf and fruit images to improve classification performance and generalization. Individual CNN-based pre-trained models, including ResNet-50, MobileNetV2, EfficientNet-B0, and ConvNeXt, were trained separately on curated datasets of mango leaf and fruit diseases. A novel Modality Attention Fusion (MAF) mechanism was introduced to dynamically weight and combine predictions from both modalities based on their discriminative strength, as some diseases are more prominent on leaves than on fruits, and vice versa. To address overfitting and improve generalization, a class-aware augmentation pipeline was integrated, which performs augmentation according to the specific characteristics of each class. The proposed attention-based fusion strategy significantly outperformed individual models and static fusion approaches, achieving a test accuracy of 99.08%, an F1 score of 99.03%, and a perfect ROC-AUC of 99.96% using EfficientNet-B0 as the base. To evaluate the model’s real-world applicability, an interactive web application was developed using the Django framework and evaluated through out-of-distribution (OOD) testing on diverse mango samples collected from public sources. These findings underline the importance of combining visual cues from multiple organs of plants and adapting model attention to contextual features for real-world agricultural diagnostics.

Producción Científica

Muhammad Mohsin mail , Muhammad Shadab Alam Hashmi mail , Irene Delgado Noya mail irene.delgado@uneatlantico.es, Helena Garay mail helena.garay@uneatlantico.es, Nagwan Abdel Samee mail , Imran Ashraf mail ,

Mohsin

en

close

Socio-economic status, food security and adherence to the Mediterranean diet in five Mediterranean countries: the DELICIOUS project

Food security is a universal need worldwide. This study explored the relationship between food security and adherence to the Mediterranean diet in the context of the DELICIOUS project. A survey involving 2,011 parents of children and adolescents aged 6–17 years was conducted. Adherence to the Mediterranean diet was assessed through the KIDMED score. Information regarding the ease of accessing Mediterranean foods, economic allowance, employment and residence was collected. Logistic regressions analyses were performed to test the associations. Individuals living in rural areas and reporting difficulty in obtaining all studied foods were less likely to follow the Mediterranean diet. Higher adherence was associated with a household monthly income higher than €4000. No associations with family status and no differences across countries were found. The progressive shift away from the Mediterranean diet may depend not only on cultural preferences for unhealthier, industrial alternatives but also on family budgets and food accessibility.

Producción Científica

Francesca Scazzina mail , Alice Rosi mail , Francesca Giampieri mail francesca.giampieri@uneatlantico.es, Carlos Poveda-Loor mail , Osama Abdelkarim mail , Mohamed Aly mail , Evelyn Frias-Toral mail , Juancho Pons mail , Laura Vázquez-Araújo mail , Sandra Sumalla Cano mail sandra.sumalla@uneatlantico.es, Iñaki Elío Pascual mail inaki.elio@uneatlantico.es, Lorenzo Monasta mail , Nadia Paladino mail , Ana Mata mail , Adrián Chacón mail , Pablo Busó mail , Giuseppe Grosso mail ,

Scazzina

<a class="ep_document_link" href="/17890/1/PIIS2001037025004581.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Enhancing detection of epileptic seizures using transfer learning and EEG brain activity signals

Epileptic seizures are neurological events characterized by sudden and excessive electrical discharges in the brain, leading to disruptions in brain function. Epileptic seizures can lead to life-threatening situations such as status epilepticus, which is characterized by prolonged or recurrent seizures and may lead to respiratory distress, aspiration pneumonia, and cardiac arrhythmias. Therefore, there is a need for an automated approach that can efficiently diagnose epileptic seizures at an early stage. The primary objective of this study is to develop a highly accurate approach for the early diagnosis of epileptic seizures. We use electroencephalography (EEG) signal data based on different brain activities to conduct experiments for epileptic seizure detection. For this purpose, a novel transfer learning technique called random forest-gated recurrent unit (RFGR) is proposed. The EEG brain activity signal data is fed into the RFGR model to generate a new feature set. The newly generated features are based on the class prediction probabilities extracted by the RFGR and are utilized to train models. Extensive experiments are carried out to investigate the performance of the proposed approach. Results demonstrate that the RFGR, when used with the random forest model, outperforms state-of-the-art techniques, achieving a high accuracy of 99.00 %. Additionally, explainable artificial intelligence analysis is utilized to provide transparent and understandable explanations of the decision-making processes of the proposed approach.

Producción Científica

Erol Kına mail , Ali Raza mail , Prudhvi Chowdary Are mail , Carmen Lilí Rodríguez Velasco mail carmen.rodriguez@uneatlantico.es, Julién Brito Ballester mail julien.brito@uneatlantico.es, Isabel de la Torre Diez mail , Naveed Anwer Butt mail , Imran Ashraf mail ,

Kına