Our study aimed to assess the role of a hand-held fundus camera and artificial intelligence (AI)-based grading system in diabetic retinopathy (DR) screening and determine its diagnostic accuracy in detecting DR compared with clinical examination and a standard fundus camera. This cross-sectional instrument validation study, as a part of the International Diabetes Federation (IDF) Diabetic Retinopathy Screening Project, included 160 patients (320 eyes) with type 2 diabetes (T2DM). After the standard indirect slit-lamp fundoscopy, each patient first underwent fundus photography with a standard 45° camera VISUCAM Zeiss and then with a hand-held camera TANG (Shanghai Zhi Tang Health Technology Co., Ltd.). Two retina specialists independently graded the images taken with the standard camera, while the images taken with the hand-held camera were graded using the DeepDR system and an independent IDF ophthalmologist. The three screening methods did not differ in detecting moderate/severe nonproliferative and proliferative DR. The area under the curve, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, kappa (ĸ) agreement, diagnostic odds ratio, and diagnostic effectiveness for a hand-held camera compared to clinical examination were 0.921, 89.1%, 100%, 100%, 91.4%, infinity, 0.11, 0.86, 936.48, and 94.9%, while compared to the standard fundus camera were 0.883, 83.2%, 100%, 100%, 87.3%, infinity, 0.17, 0.78, 574.6, and 92.2%. The results of our study suggest that fundus photography with a hand-held camera and AI-based grading system is a short, simple, and accurate method for the screening and early detection of DR, comparable to clinical examination and fundus photography with a standard camera.

Background: A proper diet plays a major role in preserving thequality of life and health for each individual. For healthcareworkers, shift work can result in irregular and poor-qualitydietary intake, which can lead to the development of healthproblems and chronic diseases.Aim: This study aimed to analyse dietary habits amonghealthcare workers based on their work schedules (morningshifts vs. rotating/shift work) using validated dietaryassessment tools.Methods: A cross-sectional study was conducted fromDecember 2023 to March 2024 in various departments ofthe Merkur Clinical Hospital. A total of 147 participants(physicians, nurses, and healthcare assistants) were surveyedusing two validated instruments: the Mediterranean DietAdherence Screener (MEDAS) and the Mediterranean DietServing Score (MDSS). Descriptive statistics and t-tests wereused to compare dietary patterns between groups, withp-values <0.05 considered statistically signifi cant.Results: A statistically signifi cant difference was observedin dessert consumption, with morning shift workersconsuming sweets more frequently (M = 3.67, SD = 2.39) thanshift workers (M = 2.70, SD = 2.21; p < 0.05). In contrast, shiftworkers consumed signifi cantly more wine than morningshift workers (M = 1.04, SD = 1.71 vs. M = 0.41, SD = 0.75;p < 0.05).Conclusion: The fi ndings suggest that work schedules areassociated with distinct dietary patterns among healthcareworkers. Morning shift workers show a preference forsweets, while shift workers consume more alcohol. Theseresults highlight the need for targeted workplace dietaryinterventions, including improved access to healthy mealsand strategies to promote balanced eating across all shifts.

Background/Objectives: Despite extensive evidence on the global burden of diabetes mellitus (DM), there is a lack of research on how sociodemographic and lifestyle factors collectively impact glycemic control and diabetes-related complications. This pilot study aimed to identify correlates of glycemic regulation, with a focus on physical activity (PA), education, and clinical variables, among adults with DM in Croatia. Methods: A cross-sectional study was conducted at the national reference center for diabetes, including 95 adult patients with type 1 or type 2 DM. Data were collected via a structured questionnaire and clinical records, covering sociodemographic characteristics, physical activity patterns, and laboratory indicators. Statistical analyses included descriptive statistics, t-tests, chi-square tests, correlation and regression analyses, and K-means clustering. Results: The mean HbA1c was 6.9% (SD = 1.3), with 33.7% of participants exceeding the 7.5% threshold. Higher education (β = −0.48, p = 0.013) and engagement in strength and balance training (β = −0.32, p = 0.041) were independently associated with lower HbA1c levels. Diabetes duration (β = 0.03, p = 0.004) and type 2 diabetes (β = 0.38, p = 0.030) were positive predictors of HbA1c. PA frequency and intensity were not significantly associated with glycemic control (p = 0.479). Cluster analysis identified three distinct profiles, with Cluster 2 (N = 10) exhibiting poor control (mean HbA1c = 9.8%) despite engaging in moderate activity. HbA1c > 7.5% was significantly associated with hypertension (OR = 3.70, 95% CI: 1.34–10.21) and retinopathy (OR = 2.82, 95% CI: 1.14–6.98). Age was the only significant predictor of cardiovascular comorbidities (OR = 1.12, p = 0.019). Education level was a predictor of participation in strength and balance training (OR = 2.45, p = 0.025). Conclusions: Glycemic control in individuals with diabetes is influenced by a combination of behavioral, clinical, and educational factors. Strength and balance training, as well as higher education, were associated with lower HbA1c levels, while older age and type 2 diabetes were indicators of a greater risk. Tailored interventions promoting structured PA and targeting educational disparities may improve clinical outcomes in this population.