DenSleepNet: DenseNet based model for sleep staging with two-frequency feature fusion and coordinate attention.

Sleep staging is often applied to assess the quality of sleep and also be used to prevent and monitor psychiatric disorders caused by sleep. However, it remains a challenge to extract the discriminative features of salient waveforms in sleep EEG and enable the network to effectively classify sleep stages by emphasizing these crucial features, thus achieving higher accuracy. In this study, an end-to-end deep learning model based on DenseNet for automatic sleep staging is designed and constructed. In the framework, two convolutional branches are devised to extract the underlying features (Two-Frequency Feature) at various frequencies, which are then fused and input into the DenseNet module to extract salient waveform features. After that, the Coordinate Attention mechanism is employed to enhance the localization of salient waveform features by emphasizing the position of salient waveforms and the spatial relationship across the entire frequency spectrum. Finally, the obtained features are accessed to the fully connected for sleep staging. The model was validated with a 20-fold cross-validation procedure on two public available datasets, and the overall accuracy, kappa coefficient, and MF1 score reached 92.9%, 78.7, 0.86 and 90.0%, 75.8, 0.80 on Sleep-EDF-20 and Sleep-EDFx, respectively. Experimental results show that the proposed model achieves competitive performance for sleep staging compared with the reported approaches under the identical conditions.

Network analysis of internet addiction and sleep disturbance symptoms.

BACKGROUND: Internet addiction (IA) is a behavioral addiction to problematic internet use. IA is associated with poorer sleep quality. Few studies to date, however, have explored the interactions between symptoms of IA and symptoms of sleep disturbance. This study uses network analysis to identify bridge symptoms by analyzing these interactions in a large sample of students. METHOD: We recruited 1977 university students to participate in our study. Each student completed the Internet Addiction Test (IAT) and the Pittsburgh Sleep Quality Index (PSQI). We used these collected data for network analysis to identify the bridge symptoms in the IAT-PSQI network by calculating the bridge centrality. Furthermore, the closest symptom connected with the bridge symptom was found to identify the comorbidity mechanisms. RESULTS: The core symptom of IA and the sleep disturbance network was "I08" (Study efficiency suffers due to internet use). The bridge symptoms between IA and sleep disturbance were "I14" (Surfing the internet late instead of sleeping), "P_DD" (Daytime dysfunction), and "I02" (Spending much time online instead of socializing in real life). Among the symptoms, "I14" had the highest bridge centrality. The edge connecting nodes "I14" and "P_SDu" (Sleep duration) had the strongest weight (0.102) around all the symptoms of sleep disturbance. Nodes "I14" and "I15" (Thinking about online shopping, games, social networking, and other network activities when unable to access the internet) had the strongest weight (0.181), connecting all the symptoms of IA. CONCLUSIONS: IA leads to poorer sleep quality, most likely by shortening sleep duration. Preoccupation with and craving the internet while being offline may lead to this situation. Healthy sleep habits should be learned, and craving may be a good point at which to treat the symptoms of IA and sleep disturbance.

Addiction Symptom Network of Young Internet Users: Network Analysis.

BACKGROUND: An increasing number of people are becoming addicted to the internet as a result of overuse. The Internet Addiction Test (IAT) is a popular tool for evaluating internet use behaviors. The interaction between different symptoms and the relationship between IAT and clinical diagnostic criteria are not well understood. OBJECTIVE: This study aimed to explore the core symptoms of internet addiction (IA) and the correlation between different symptoms of the IA symptom network. Network analysis was also conducted to explore the association between the IAT scale and the Diagnostic and Statistical Manual of Mental Disorders-5th edition (DSM-5) criteria for IA. METHODS: We recruited 4480 internet users (aged 14-24 years), and they completed the IAT. The final analysis included 63.50% (2845/4480) of the participants after screening the submitted questionnaires. Participants were classified into IA group and non-IA (NIA) group. By using partial correlation with Lasso regularization networks, we identified the core symptoms of IA in each group and compared the group differences in network properties (strength, closeness, and betweenness). Then, we analyzed the symptom networks of the DSM-5 diagnostic criteria and IAT scale for IA. RESULTS: A total of 12.47% (355/2845) of the patients were in the IA group and 87.52% (2490/2845) of the patients were in the NIA group, and both groups were evaluated for the following nodes: IAT_06 (school work suffers; strength=0.511), IAT_08 (job performance suffers; strength=0.531), IAT_15 (fantasize about being on the web; strength=0.474), IAT_17 (fail to stop being on the web; strength=0.526), and IAT_12 (fear about boredom if offline; strength=0.502). The IA groups had a stronger edge between IAT_09 (defensive or secretive about being on the web) and IAT_18 (hidden web time) than the NIA groups. The items in DSM-5 had a strong association with IAT_12 (weight=-0.066), IAT_15 (weight=-0.081), IAT_17 (weight=-0.106), IAT_09 (weight=-0.198), and IAT_18 (weight=-0.052). CONCLUSIONS: The internet use symptom network of the IA group is significantly different from that of the NIA group. Nodes IAT_06 (school work affected) and IAT_08 (work performance affected) are the resulting symptoms affected by other symptoms, whereas nodes IAT_12 (fear about boredom if offline), IAT_17 (inability to stop being on the web), and IAT_15 (fantasize about being on the web) are key symptoms that activate other symptoms of IA and are strongly linked to the inability to control the intention to play games in the DSM-5.

The relationship between teacher professional identity and burnout amid the pandemic: A moderated mediation model.

Background: Teacher burnout is affected by personal and social factors. COVID-19 has greatly impacted teachers' physical and mental health, which could aggravate teacher burnout. Purpose: Based on the JD-R model, this study aims to investigate the relationship between teacher professional identity (TPI) and job burnout during the COVID-19 pandemic, and examine the moderating roles of perceived organizational support (POS) and psychological resilience (PR) in these relationships among primary and secondary school teachers in China. Methods: A total of 3,147 primary and secondary school teachers participated in this study. Findings: Work engagement played a mediating role in the relationship between professional identity and burnout; when the POS and PR scores were high, the predictive coefficient of TPI on burnout was the largest. Originality: This study tested the mechanism underlying the relationship between TPI and burnout, and explored the protective factors of burnout. Implications: This study supports the applicability of the JD-R model during COVID-19 and provides ideas for teachers to reduce burnout.