Efficient Multi-View K-Means for Image Clustering.

Nowadays, data in the real world often comes from multiple sources, but most existing multi-view K-Means perform poorly on linearly non-separable data and require initializing the cluster centers and calculating the mean, which causes the results to be unstable and sensitive to outliers. This paper proposes an efficient multi-view K-Means to solve the above-mentioned issues. Specifically, our model avoids the initialization and computation of clusters centroid of data. Additionally, our model use the Butterworth filters function to transform the adjacency matrix into a distance matrix, which makes the model is capable of handling linearly inseparable data and insensitive to outliers. To exploit the consistency and complementarity across multiple views, our model constructs a third tensor composed of discrete index matrices of different views and minimizes the tensor's rank by tensor Schatten p-norm. Experiments on two artificial datasets verify the superiority of our model on linearly inseparable data, and experiments on several benchmark datasets illustrate the performance.

The effect of triglycerides in the associations between physical activity, sedentary behavior and depression: An interaction and mediation analysis.

BACKGROUND: Previous studies reported that there were associations between physical activity (PA), sedentary behavior (SB), triglycerides (TG) and depression. However, it is unclear whether there is an interaction and/or mediation effect of TG in the associations between PA, SB and depression. This study was to explore the effects of TG in the associations between PA, SB and depression, respectively. METHODS: Our study included 7,728 adults (aged ≥20 years) participating in the 2009-2018 National Health and Nutrition Examination Survey (NHANES), which was a nationally representative survey. With higher TG as a mediator, a four-way decomposition method was used to decompose the total effects of insufficient PA and prolonged SB on depression, respectively. RESULTS: There were statistically significant associations between moderate-to-vigorous recreational activity (MVRA), SB, TG and depression (all P < 0.05), respectively. With higher TG as a mediator, there were interaction and mediation effects in the association between SB and depression, and the excess relative risk (95% CI) of reference interaction effect (INTref) and the total mediation effect were 0.15(0.01,0.29) and 0.03(0.00,0.06), respectively. While insufficient MVRA attributed a proportion of 80.42% for the controlled direct effect (CDE) and 6.28% for total mediation effect. In addition, similar findings were observed in those with hypertension, overweight/obesity, dyslipidemia, and metabolic syndromes. LIMITATION: This was a cross-sectional study, and the ability to make causal inference was limited. CONCLUSION: The controlled direct effect of MVRA, the mediation of TG, and the interaction between SB and TG played the vital roles in increasing the risk of depression.

The mediating and interacting role of physical activity and sedentary behavior between diabetes and depression in people with obesity in United States.

AIMS: Sufficient physical activity (PA) and less sedentary behavior (SB) have antidepressant effects, however the amount of PA varies according to weight status in adults with diabetes. Given that depression is a common complication of diabetes, we aimed to quantify to what extent the effects of diabetes on the risk of depression were explained by SB or insufficient PA in adults with and without obesity. METHODS: Data were collected from the 2007-2016 National Health and Nutrition Examination Survey (NHANES), which included 22,304 participants aged 20 years or older. The mediation and interaction were evaluated by the four-way effect decomposition method. RESULTS: In individuals suffering from obesity, diabetes, SB, and insufficient moderate and vigorous intensity recreational activity were associated with an increased risk of depression. The increased relative risk of diabetes on depression could be decomposed into a reference interaction (96.78%), a mediated interaction (5.07%), and a pure indirect effect (2.48%), with insufficient moderate-intensity recreational activity as a potential mediator. SB attributed a proportion of 89.21% for the controlled direct effect and 3.64% for pure indirect effect. CONCLUSION: Adults with obesity and diabetes are encouraged to increase moderate-intensity recreational activity and reduce SB, which can greatly prevent the occurrence of depression.

Association of smartphone addiction with daily behaviors and mental health during the COVID-19 pandemic among medical college students / 中国学校卫生

Objective@#To investigate the smartphone addiction among college students during the Coronavirus Disease 2019 (COVID-19) pandemic and its association with daily behaviors and mental health,and to provide reference for heath education and psychological counseling for college students.@*Methods@#An observational study using online quyestionnaire was conducted among 10 357 college students of two provincial medical schools in Guangdong and Shanxi Province from February 24th to March 4th in 2020. Participants were investigated on demographic information, smartphone addiction, daily routine, physical activity, weight status, anxiety, and other health information. Logistic regression with inverse probability treatment weighting (IPTW) based on propensity score was used to analyze the association between smartphone addiction with daily behavior and mental health.@*Results@#The prevalence of smartphone addiction was 59.42%. The prevalence of phone addiction was higher in postgraduates, senior undergraduates, students with non-medical major, students living in GuangDong and those without regular exercise habit before vacation(χ 2=47.91，17.78，42.75，138.58，P<0.05). With IPTW, there were significant associations between smartphone addiction and late bedtimes (OR=1.82, 95%CI=1.66-1.98) and wake-up times (OR=1.55, 95%CI=1.44-1.68), more sedentary behaviors (OR=1.21, 95%CI=1.12-1.31), less moderate to vigorous physical activity (OR=1.33, 95%CI=1.22-1.44), anxiety (OR=2.98, 95%CI=2.52-3.40), weight gain(OR=1.27，95%CI=1.17-1.37) and other detrimental daily behavior and feelings.@*Conclusion@#High prevalence of smartphone addiction has been observed during the COVID-19 pandemic, with impaired daily behavior and mental health.

The effects of physical activity and sedentary behavior in the associations between cardiovascular diseases and depression: A four-way decomposition.

BACKGROUND: Negative effects of cardiovascular diseases (CVDs) on depression have been reported, but the relative contribution of physical activity (PA) and sedentary behavior (SB) to such effects remains unclear. Our objective was to precisely quantify the effects modified or mediated by PA and SB using the recently developed four-way effect decomposition. METHODS: Our analysis included 22,117 adults (aged≥20 years) participating in the 2007-2016 National Health and Nutrition Examination Survey (NHANES). Depression was assessed by the Patient Health Questionnaire-9. Moderate recreational activity (MRA), moderate work activity (MWA), and walk or bicycle for transportation, were measured by the Global Physical Activity Questionnaire (GPAQ). A four-way decomposition was applied with PA and SB as mediator to decompose the total effects of CVDs on depression into four components: controlled direct effect (CDE), pure indirect effect (PIE), reference interaction (INTref), and mediated interaction effect (INTmed). RESULTS: There were statistically significant associations between CVDs, MRA, and depression, with the adjusted odds ratios (95% confidence intervals) of being 1.73 (1.42~2.30), 1.81 (1.42~2.30), respectively. With MRA as a mediator, there were interaction and mediation effects in the associations between CVDs and depression, and the proportions attributable of INTref, INTmed, and PIE were 94.08% (P<0.001), 3.92% (P = 0.007), and 1.68% (P = 0.006), respectively. Furthermore, the mediation effect was statistically significant in females rather than males. CONCLUSIONS: MRA strongly modified and mediated the effects of CVDs on depression, especially in females. Our results suggested that sufficient MRA (at least 150 min per week) was considered as requirement for preventing depression in CVDs patients in females.

Efficacy and Safety of Direct Oral Anticoagulants in Elderly Patients With Atrial Fibrillation: A Network Meta-Analysis.

Background: Direct oral anticoagulants (DOACs) have been widely used in patients with atrial fibrillation (AF) for antithrombotic prophylaxis, which were shown to have a favorable risk-benefit profile. However, there are no guidelines for the use of DOACs in elderly patients (aged ≥75 years) with AF, which creates uncertainty about the optimal antithrombotic treatment in these patients. Methods: After comprehensively searching Embase, PubMed, and Cochrane databases, five phase III randomized controlled trials involving 28,137 elderly participants were included in this study. The efficacy outcome was stroke or systemic embolism, and the safety outcome was major bleeding. We conducted a network meta-analysis by using a Bayesian random-effect model for the first time to evaluate the efficacy and safety of main DOACs (apixaban, edoxaban, rivaroxaban, and dabigatran) and warfarin in elderly patients with AF. Hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) were used to assess the effect of drugs on efficacy and safety. The rank probabilities were used to reflect the hierarchy of drugs, and a larger rank probability value symbolized a better rank of drugs. Results: In the prophylaxis of stroke or systemic embolism, apixaban was found to be the best among DOACs compared to warfarin (HR, 0.71; 95% CI: 0.33-1.50), though this finding was not statistically significant. Apixaban ranked the best (rank probabilities, 41.2%) in efficacy of drugs, followed by rivaroxaban, edoxaban, dabigatran, and warfarin (rank probabilities, 31.8, 15.9, 10.9, and 0.2%, respectively). In reducing the risk of major bleeding, apixaban was found to be the best among DOACs too, compared to warfarin (HR, 0.64; 95% CI: 0.33-1.30), though this finding was not statistically significant. In safety, apixaban ranked the best (rank probabilities, 71.4%), followed by edoxaban, dabigatran, warfarin, and rivaroxaban (rank probabilities, 21.0, 5.8, 0.9, and 0.8%, respectively). Conclusions: DOACs showed a lower incidence of stroke/systemic embolism and major bleeding compared with warfarin in antithrombotic therapy in elderly patients (aged ≥75 years), with apixaban being the best of those interventions. Therefore, apixaban should be given priority as an anticoagulant in stroke prevention for elderly patients with AF.