Background: The association between visceral adiposity index (VAI) and diabetic complications has been reported in cross-sectional studies, while the effect of VAI on complication development remains unclear. This study aims to evaluate the longitudinal association of VAI and Chinese VAI (CVAI) with the incidence of diabetic nephropathy and retinopathy using a Chinese cohort. Methods: A total of 8 948 participants with type 2 diabetes from Beijing Health Management Cohort were enrolled during 2013–2014, and followed until December 31, 2019. Nephropathy was confirmed by urine albumin/creatinine ratio and estimated glomerular filtration rate; retinopathy was diagnosed using fundus photograph. Results: The mean (SD) age was 53.35 (14.66) years, and 6 154 (68.8%) were men. During a median follow-up of 4.82 years, 467 participants developed nephropathy and 90 participants developed retinopathy. One-SD increase in VAI and CVAI levels were significantly associated with an increased risk of nephropathy, and the adjusted hazard ratios (HR) were 1.127 (95% CI 1.050–1.210) and 1.165 (95% CI 1.003–1.353), respectively. On contrary, VAI and CVAI level were not associated with retinopathy after adjusting confounding factors. Conclusion: VAI and CVAI are independently associated with the development of nephropathy, but not retinopathy in Chinese adults with diabetes.

Background: To evaluate the association of physical activity (PA) intensity with cognitive performance at baseline and during follow-up. Methods: A total of 4039 participants aged 45 years or above from the China Health and Retirement Longitudinal Study were enrolled in visit 1 (2011–2012) and followed for cognitive function in visit 2 (2013–2014), visit 3 (2015–2016), and visit 4 (2017–2018). We analyzed the association of PA intensity with global cognition, episodic memory, and mental intactness at baseline using adjusted regression methods and evaluated the long-term effect of PA intensity using multiple measures of cognition scores by mixed effect model. Results: In cross-sectional analysis, mild and moderate PA, rather than vigorous PA, was associated with better cognitive performance. The results remained consistent in multiple sensitivity analyses. During the follow-up, participant with mild PA had a 0.56 (95% CI 0.12–0.99) higher global cognition, 0.23 (95% CI 0.01–0.46) higher episodic memory, and 0.33 (95% CI 0.01–0.64) higher mental intactness, while those with moderate PA had a 0.74 (95% CI 0.32–1.17) higher global score, 0.32 (95% CI 0.09–0.54) higher episodic memory, and 0.43 (95% CI 0.12–0.74) higher mental intactness, compared with individuals without PA. Vigorous PA was not beneficial to the long-term cognitive performance. Conclusions: Our study indicates that mild and moderate PA could improve cognitive performance, rather than the vigorous activity. The targeted intensity of PA might be more effective to achieve the greatest cognition improvement considering age and depressive status.

Background: The supervised deep learning approach provides state-of-the-art performance in a variety of fundus image classification tasks, but it is not applicable for screening tasks with numerous or unknown disease types. The unsupervised anomaly detection (AD) approach, which needs only normal samples to develop a model, may be a workable and cost-saving method of screening for ocular diseases. Objective: This study aimed to develop and evaluate an AD model for detecting ocular diseases on the basis of color fundus images. Methods: A generative adversarial network–based AD method for detecting possible ocular diseases was developed and evaluated using 90,499 retinal fundus images derived from 4 large-scale real-world data sets. Four other independent external test sets were used for external testing and further analysis of the model’s performance in detecting 6 common ocular diseases (diabetic retinopathy [DR], glaucoma, cataract, age-related macular degeneration, hypertensive retinopathy [HR], and myopia), DR of different severity levels, and 36 categories of abnormal fundus images. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the model’s performance were calculated and presented. Results: Our model achieved an AUC of 0.896 with 82.69% sensitivity and 82.63% specificity in detecting abnormal fundus images in the internal test set, and it achieved an AUC of 0.900 with 83.25% sensitivity and 85.19% specificity in 1 external proprietary data set. In the detection of 6 common ocular diseases, the AUCs for DR, glaucoma, cataract, AMD, HR, and myopia were 0.891, 0.916, 0.912, 0.867, 0.895, and 0.961, respectively. Moreover, the AD model had an AUC of 0.868 for detecting any DR, 0.908 for detecting referable DR, and 0.926 for detecting vision-threatening DR. Conclusions: The AD approach achieved high sensitivity and specificity in detecting ocular diseases on the basis of fundus images, which implies that this model might be an efficient and economical tool for optimizing current clinical pathways for ophthalmologists. Future studies are required to evaluate the practical applicability of the AD approach in ocular disease screening.

Research on the relationship between metabolic syndrome (MetS), its components and eating speed is limited in China. The present study aimed to clarify the association between MetS, its components and eating speed in a Beijing adult population. This cross-sectional study included 7972 adults who were 18–65 years old and who received health check-ups at the Beijing Physical Examination Center in 2016. Logistic regression was conducted to explore the associations between MetS, its components and eating speed. The prevalence of MetS in this population was 24.65% (36.02% for males and 10.18% for females). Eating speed was significantly associated with a high risk for MetS, elevated blood pressure, and central obesity for both genders. Eating speed was associated with a high risk for elevated triglycerides and with a reduction in high-density lipoprotein in males, and eating speed was associated with a high risk for elevated fasting plasma glucose in females. Compared with slow eating speed, the multivariate-adjusted odds ratios of medium eating speed and fast eating speed for MetS were 1.65 (95% confidence interval 1.32–2.07) and 2.27 (95% confidence interval 1.80–2.86) for all subjects, 1.58 (95% confidence interval 1.21–2.07) and 2.21 (95% confidence interval 1.69–2.91) for males, and 1.75 (95% confidence interval 1.15–2.68) and 2.27 (95% confidence interval 1.46–3.53) for females, respectively. Eating speed is positively associated with MetS and its components. Future recommendations aiming to prevent MetS and its components may focus on eating speed.