Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

Dipòsit Digital de Documents de la UAB
Publication Artículo científico (JournalArticle)

Mediterranean Diet Is a Predictor of Progression of Subclinical Atherosclerosis in a Mediterranean Population, The ILERVAS Prospective Cohort Study

  • Rojo López, Marina Idalia|||0000-0001-6748-3156
  • Bermúdez-López, Marcelino|||0000-0002-3188-4158
  • Castro, Eva
  • Farràs-Sallés, Cristina|||0000-0001-5956-4116
  • Torres, Gerard|||0000-0002-1417-7320
  • Pamplona, Reinald|||0000-0003-4337-6107
  • Lecube, Albert|||0000-0001-9684-0183
  • Valdivieso, José Manuel
  • Fernández, Elvira|||0000-0001-5236-1762
  • Julve i Gil, Josep|||0000-0002-6531-2246
  • Castelblanco, Esmeralda|||0000-0002-2061-6270
  • Alonso Pedrol, Núria|||0000-0001-9687-8050
  • Antentas Peraile, María|||0009-0005-8140-203X
  • Barranco-Altirriba, Maria|||0000-0002-1525-210X
  • Perera Lluna, Alexandre|||0000-0001-6427-851X
  • Franch-Nadal, Josep|||0000-0002-5175-1555
  • Granado-Casas, Minerva|||0000-0003-3447-5728
  • Mauricio, Didac|||0000-0002-2868-0250
Atherosclerotic cardiovascular disease remains a major health issue, often developing silently as subclinical atherosclerotic disease (SAD). The Mediterranean diet (MDiet) is known for its cardiovascular benefits, but the combined influence of both MDiet adherence and physical activity (PA) on SAD progression has not been previously documented. Objective: We aimed to investigate how adherence to a healthy lifestyle, defined as MDiet adherence and PA level, influences SAD progression in subjects from the ILERVAS cohort follow-up. Methods: A study on 3097 participants from the ILERVAS prospective cohort was conducted. MDiet adherence was assessed using the MEDAS score, and PA categories were established using the IPAQ, both categorized into low, moderate, and high levels. Two different lifestyle scores integrating the MDiet and PA categories were built. The presence of atherosclerotic plaques was assessed by carotid and femoral ultrasound examination. Demographic, clinical, and biochemical data were also obtained. Multivariable linear, logistic, and Poisson regression models adjusted for potential confounders were used to analyze the association between the lifestyle scores and SAD progression, as well as the MDiet and PA as separate variables and number of territories with plaque. Results: A healthier lifestyle score did not show an effect on SAD progression. However, a higher MEDAS score was associated with a 3% decrease in the number of territories with plaque (IRR 0.97, 95% CI 0.96-0.99, p < 0.001), suggesting a protective effect of the adherence to the MDiet. PA did not show a significant association (IRR 1.00, 95% CI 1.00-1.00, p = 0.269). Older age, hypertension, dyslipidemia, smoking, and lower eGFR were associated with SAD progression, while the female sex was protective (IRR 0.67, 95% CI 0.63-0.72, p < 0.001). Conclusions: The findings of this study show that higher adherence to the MDiet is associated with reduced incidence of SAD, indicating its potential role in cardiovascular prevention strategies. Although a higher lifestyle score or physical activity levels did not show any significant effect, promoting the MDiet, alongside managing traditional cardiovascular risk factors, could be an effective public health intervention to prevent atherosclerosis and reduce the burden of cardiovascular disease.
Dipòsit Digital de Documents de la UAB
Publication Artículo científico (JournalArticle)

A deep attention-based encoder for the prediction of type 2 diabetes longitudinal outcomes from routinely collected health care data

  • Manzini, E.|||0000-0003-2961-5335
  • Vlacho, Bogdan|||0000-0003-3768-8112
  • Franch-Nadal, Josep|||0000-0002-5175-1555
  • Escudero, J.
  • Génova, A.
  • Reixach, Elisenda|||0000-0001-8557-0771
  • Andrés, E.
  • Pizarro, I.
  • Mauricio, Didac|||0000-0002-2868-0250
  • Perera Lluna, Alexandre|||0000-0001-6427-851X
Recent evidence indicates that Type 2 Diabetes Mellitus (T2DM) is a complex and highly heterogeneous disease involving various pathophysiological and genetic pathways, which presents clinicians with challenges in disease management. While deep learning models have made significant progress in helping practitioners manage T2DM treatments, several important limitations persist. In this paper we propose DARE, a model based on the transformer encoder, designed for analyzing longitudinal heterogeneous diabetes data. The model can be easily fine-tuned for various clinical prediction tasks, enabling a computational approach to assist clinicians in the management of the disease. We trained DARE using data from over 200,000 diabetic subjects from the primary healthcare SIDIAP database, which includes diagnosis and drug codes, along with various clinical and analytical measurements. After an unsupervised pre-training phase, we fine-tuned the model for predicting three specific clinical outcomes: i) occurrence of comorbidity, ii) achievement of target glycemic control (defined as glycated hemoglobin <7%) and iii) changes in glucose-lowering treatment. In cross-validation, the embedding vectors generated by DARE outperformed those from baseline models (comorbidities prediction task AUC=0.88, treatment prediction task AUC=0.91, HbA1c target prediction task AUC=0.82). Our findings suggest that attention-based encoders improve results with respect to different deep learning and classical baseline models when used to predict different clinical relevant outcomes from T2DM longitudinal data.