Recently, online education has been gaining prominence in university life. Our survey aimed to examine sleep quality and perceived stress levels among students at the University of Pécs Faculty of Health Sciences. A cross-sectional, quantitative, descriptive survey was conducted between February and March 2023. The online survey included the Hungarian versions of the internationally validated Athens Insomnia Scale (AIS) and Perceived Stress Scale (PSS). Statistical analysis involved descriptive statistics, independent t-tests, analysis of variance (ANOVA), and Mann–Whitney and Kruskal–Wallis tests (p < 0.05). We analyzed 304 responses, and females dominated (n = 270; 88.8%). Students in a relationship had significantly higher AIS scores (t = −2.470; p = 0.014). Medium average (2.50–3.49) students and those who rarely/never exercise showed significantly higher AIS and PSS (p ≤ 0.05). Students on the phone/watching a series during online education, daily laptop/TV use for more than 2 h, and pre-sleep use of smart devices for more than 60 min also negatively affected AIS and PSS scores (p ≤ 0.05). Nursing, physiotherapy, and radiography students were the most affected regarding insomnia and perceived stress (p ≤ 0.05). Our survey shows that excessive smart device use and lack of exercise are associated with higher stress levels and poorer sleep quality.

Maintaining mental health is essential for professions with higher stress levels and challenging environments, including emergency specializations. In this study, the occurrence of distress, anxiety, and depression among a group of ambulance and hospital emergency care professionals was assessed (n = 202). A cross-sectional, quantitative, descriptive online survey was conducted, including the internationally validated Beck depression inventory (BDI), the perceived stress scale (PSS-14), and the State–Trait Anxiety Inventory (STAI). Statistical analyses involved descriptive statistics, the χ²-test, Mann–Whitney U test, Kruskal–Wallis test, Dunn–Bonferroni test, logistic regression (LR), Cramer coefficient (Cramer’s V), Kolmogorov–Smirnov test, and Spearman’s rank correlation coefficient (r_s). Based on the results, female professionals are more likely to have depressive symptoms (OR = 2.6, 95% CI = 1.3–5.1), perceived stress (OR = 1.2, 95% CI = 1.2–4.1), and anxiety (OR = 2.1, 95% CI = 1.0–4.1) than male professionals. Perceived stress levels decreased proportionally with increasing years spent working in healthcare (OR = 7.4, 95% CI = 7.1–8.3). Extended work shifts of 12 or 24 h increase the risk of perceived stress and anxiety in emergency care workers (p = 0.02). Customized stress management interventions are needed to mitigate the amplified mental strain associated with gender, working years, and longer shifts in the emergency care sector to sustain their mental health and well-being.

In complex and dynamic educational environments, the discovery of categories of teacher instructional behavior indicators is crucial for teaching evaluation. Generalized Category Discovery (GCD) offers a novel approach for automatically uncovering such indicators from teaching videos. However, existing methods suffer from inadequate capture of spatiotemporal dynamics and semantic interference in complex classroom environments, which limits the effectiveness of novel category discovery. This paper proposes the MGCL model, a multi-granularity contrastive learning framework. Guided by known indicator category data, it mines large-scale unlabeled data to achieve automated identification and discovery of new categories for instructional behavior indicators. In this framework, instance-level contrastive learning (IL-CL) optimizes the characteristic representation of individual samples; neighborhood consistency contrastive learning (NC-CL) enhances the discriminability of samples within local contexts by emphasizing the semantic consistency of neighboring samples; and global category consolidation contrastive learning (GC-CL) leverages pseudo-labels to optimize category partitioning at a holistic level, thereby improving inter-class separation. Operating at the instance, neighborhood, and global levels respectively, these components work in concert to collectively enhance feature discriminability and category stability. Experiments were conducted on the proprietary TBU dataset and the public UCF101 and HMDB51 datasets. Results demonstrate that the proposed method achieves state-of-the-art performance across multiple metrics. This study provides a feasible and effective solution for discovering categories of instructional behavior indicators in complex educational environments.