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)

Machine learning for anxiety and depression profiling and risk assessment in the aftermath of an emergency

  • Villanueva Benito, Guillermo
  • Goldberg, Ximena|||0000-0001-9681-5826
  • Brachowicz, Nicolai
  • Castaño-Vinyals, Gemma|||0000-0003-4468-1816
  • Blay, Natalia|||0000-0002-7160-8529
  • Espinosa, Ana|||0000-0001-7399-3267
  • Davidhi, Flavia
  • Torres, Diego
  • Kogevinas, Manolis|||0000-0002-9605-0461
  • de Cid, Rafael|||0000-0003-3579-6777
  • Petrone, Paula Marcela
Background & objectives: Mental health disorders pose an increasing public health challenge worsened by the COVID-19 pandemic. The pandemic highlighted gaps in preparedness, emphasizing the need for early identification of at-risk groups and targeted interventions. This study aims to develop a risk assessment tool for anxiety, depression, and self-perceived stress using machine learning (ML) and explainable AI to identify key risk factors and stratify the population into meaningful risk profiles. Methods: We utilized a cohort of 9291 individuals from Northern Spain, with extensive post-COVID-19 mental health surveys. ML classification algorithms predicted depression, anxiety, and self-reported stress in three classes: healthy, mild, and severe outcomes. A novel combination of SHAP (SHapley Additive exPlanations) and UMAP (Uniform Manifold Approximation and Projection) was employed to interpret model predictions and facilitate the identification of high-risk phenotypic clusters. Results: The mean macro-averaged one-vs-one AUROC was 0.77 (± 0.01) for depression, 0.72 (± 0.01) for anxiety, and 0.73 (± 0.02) for self-perceived stress. Key risk factors included poor self-reported health, chronic mental health conditions, and poor social support. High-risk profiles, such as women with reduced sleep hours, were identified for self-perceived stress. Binary classification of healthy vs. at-risk classes yielded F1-Scores over 0.70. Conclusions: Combining SHAP with UMAP for risk profile stratification offers valuable insights for developing effective interventions and shaping public health policies. This data-driven approach to mental health preparedness, when validated in real-world scenarios, can significantly address the mental health impact of public health crises like COVID-19.
Dipòsit Digital de Documents de la UAB
Publication Artículo científico (JournalArticle)

Sex-specific chrono-nutritional patterns and association with body weight in a general population in Spain (GCAT study)

  • Pons-Muzzo, Luciana
  • de Cid, Rafael|||0000-0003-3579-6777
  • Obón-Santacana, Mireia|||0000-0003-4646-3513
  • Straif, Kurt
  • Papantoniou, Kyriaki|||0000-0002-0340-9659
  • Santonja, Isabel|||0000-0002-6736-7988
  • Kogevinas, Manolis|||0000-0002-9605-0461
  • Palomar-Cros, Anna|||0000-0001-8151-9290
  • Lassale, Camille|||0000-0002-9340-2708
Background: Altered meal timing patterns can disrupt the circadian system and affect metabolism. Our aim was to describe sex-specific chrono-nutritional patterns, assess their association with body mass index (BMI) and investigate the role of sleep in this relationship. Methods: We used the 2018 questionnaire data from the population-based Genomes for Life (GCAT) (n = 7074) cohort of adults aged 40-65 in Catalonia, Spain, for cross-sectional analysis and its follow-up questionnaire data in 2023 (n = 3128) for longitudinal analysis. We conducted multivariate linear regressions to explore the association between mutually adjusted meal-timing variables (time of first meal, number of eating occasions, nighttime fasting duration) and BMI, accounting for sleep duration and quality, and additional relevant confounders including adherence to a Mediterranean diet. Finally, cluster analysis was performed to identify chrono-nutritional patterns, separately for men and women, and sociodemographic and lifestyle characteristics were compared across clusters and analyzed for associations with BMI. Results: In the cross-sectional analysis, a later time of first meal (β 1 h increase = 0.32, 95% CI 0.18, 0.47) and more eating occasions (only in women, β 1 more eating occasion = 0.25, 95% CI 0.00, 0.51) were associated with a higher BMI, while longer nighttime fasting duration with a lower BMI (β 1 h increase=-0.27, 95% CI -0.41, -0.13). These associations were particularly evident in premenopausal women. Longitudinal analyses corroborated the associations with time of first meal and nighttime fasting duration, particularly in men. Finally, we obtained 3 sex-specific clusters, that mostly differed in number of eating occasions and time of first meal. Clusters defined by a late first meal displayed lower education and higher unemployment in men, as well as higher BMI for both sexes. A clear "breakfast skipping" pattern was identified only in the smallest cluster in men. Conclusions: In a population-based cohort of adults in Catalonia, we found that a later time of first meal was associated with higher BMI, while longer nighttime fasting duration associated with a lower BMI, both in cross-sectional and longitudinal analyses.