Background: Recent studies offer conflicting conclusions about the effectiveness of digital health interventions in changing physical activity behaviors. In addition, research focusing on digital health interventions for college students remains relatively scarce. Objective: This study aims to examine the impact of digital health interventions on physical activity behaviors among college students, using objective measures as outcome indicators. Methods: In accordance with the 2020 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, a comprehensive literature search was conducted across several databases, including MEDLINE (PubMed), Web of Science, Cochrane Library, and EBSCO (CINAHL Plus with full text), to identify relevant intervention studies published up to June 6, 2023. The inclusion criteria specified studies that examined the quantitative relationships between digital health interventions and physical activity among adults aged 18 years to 29 years, focusing on light physical activity (LPA), moderate to vigorous physical activity (MVPA), sedentary time (ST), or steps. Non–randomized controlled trials were excluded. The quality of the studies was assessed using the Cochrane Risk of Bias tool. Results were synthesized both narratively and quantitatively, where applicable. When sufficient homogeneity was found among studies, a random-effects model was used for meta-analysis to account for variability. Results: In total, 8 studies, encompassing 569 participants, were included in the analysis. The primary outcomes measured were LPA, MVPA, ST, and steps. Among these studies, 3 reported on LPA, 5 on MVPA, 5 on ST, and 3 on steps. The meta-analysis revealed a significant increase in steps for the intervention group compared with the control group (standardized mean difference [SMD] 0.64, 95% CI 0.37-0.92; P<.001). However, no significant differences were observed between the intervention and control groups regarding LPA (SMD –0.08, 95% CI –0.32 to 0.16; P=.51), MVPA (SMD 0.02, 95% CI –0.19 to 0.22; P=.88), and ST (SMD 0.03, 95% CI –0.18 to 0.24; P=.78). Conclusions: Digital health interventions are effective in increasing steps among college students; however, their effects on LPA, MVPA, and sedentary behavior are limited. Trial Registration: PROSPERO CRD42024533180; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=533180

Background: The exponential growth of digital technologies and the ubiquity of social media platforms have led to unprecedented mental health challenges among college students, highlighting the critical need for effective intervention approaches. Objective: This study aimed to explore the relationship between meeting the 24-hour movement guidelines (24-HMG) health behavior combinations and the risk of social network addiction (SNA) as well as mental health issues among university students. It further sought to compare differences in mental health indicators and SNA levels across various risk groups and adherence patterns, and to identify the optimal 24-HMG health behavior intervention strategies for students at high risk of SNA. Methods: This cross-sectional study recruited a total of 12,541 university students from the university town of Guizhou Province as participants. Data were collected through standardized questionnaires, including the Chinese version of Social Network Addiction Scale for College Students (SNAS-C), the adult attention-deficit/hyperactivity disorder (ADHD) self-report scale (ASRS), and the Chinese version of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) Self-Report Level 1 Cross-Cutting Symptom Measure for Adults (DSM-5 CCSM), among others. The primary analytical method used was the random forest model, which was used to explore the relationship between different 24-HMG behavior combinations and mental health variables among student at high-risk of SNA. In addition, the study aimed to identify the optimal 24-HMG health behavior intervention strategies for this high-risk group. Results: Participants in the meeting none group exhibited the highest SNA scores (57.98), which declined progressively with greater adherence. Among single-guideline groups, meeting physical activity (PA; 53.07) and meeting sedentary time (ST; 52.72) showed similar scores. Further reductions were seen in meeting PA+ST (49.68), meeting sleep (48.44), and meeting ST+sleep (44.75), with the lowest in meeting PA+ST+sleep. Approximately 6% of the variance in SNA was attributable to differences in adherence patterns (η²=0.06). Students meeting all three 24-HMG components—PA, sleep, and ST—demonstrated the strongest protection against attention deficit, depression, and anxiety. All 24-HMG behaviors were inversely associated with mental health symptoms, except academic satisfaction, which was positively correlated. Random forest modeling identified meeting sleep+ST as the most impactful for mania (0.4491), sleep disturbance (0.4032), personality (0.3924), and dissociation (0.3832). Meeting ST alone showed the strongest effects on substance (0.6176) and alcohol use (0.6597). Depression was influenced by meeting sleep+ST (0.2053), meeting PA+ST+sleep (0.1650), and meeting PA+ST (0.1634). The model achieved high accuracy for ASRS (0.912; F₁-score=0.927), with robust predictions for substance use (F₁-score=0.873) and mania (F₁-score=0.836). Conclusions: Adherence to the health behaviors recommended by the 24-HMG can significantly improve the mental health outcomes of university students at high risk for SNA. The findings of this study support the development of mental health intervention strategies for students at high-risk of SNA based on the 24-HMG framework.

BACKGROUND Recent studies offer conflicting conclusions about the effectiveness of digital health interventions in changing physical activity behaviors. In addition, research focusing on digital health interventions for college students remains relatively scarce. OBJECTIVE This study aims to examine the impact of digital health interventions on physical activity behaviors among college students, using objective measures as outcome indicators. METHODS In accordance with the 2020 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, a comprehensive literature search was conducted across several databases, including MEDLINE (PubMed), Web of Science, Cochrane Library, and EBSCO (CINAHL Plus with full text), to identify relevant intervention studies published up to June 6, 2023. The inclusion criteria specified studies that examined the quantitative relationships between digital health interventions and physical activity among adults aged 18 years to 29 years, focusing on light physical activity (LPA), moderate to vigorous physical activity (MVPA), sedentary time (ST), or steps. Non–randomized controlled trials were excluded. The quality of the studies was assessed using the Cochrane Risk of Bias tool. Results were synthesized both narratively and quantitatively, where applicable. When sufficient homogeneity was found among studies, a random-effects model was used for meta-analysis to account for variability. RESULTS In total, 8 studies, encompassing 569 participants, were included in the analysis. The primary outcomes measured were LPA, MVPA, ST, and steps. Among these studies, 3 reported on LPA, 5 on MVPA, 5 on ST, and 3 on steps. The meta-analysis revealed a significant increase in steps for the intervention group compared with the control group (standardized mean difference [SMD] 0.64, 95% CI 0.37-0.92; P<.001). However, no significant differences were observed between the intervention and control groups regarding LPA (SMD –0.08, 95% CI –0.32 to 0.16; P=.51), MVPA (SMD 0.02, 95% CI –0.19 to 0.22; P=.88), and ST (SMD 0.03, 95% CI –0.18 to 0.24; P=.78). CONCLUSIONS Digital health interventions are effective in increasing steps among college students; however, their effects on LPA, MVPA, and sedentary behavior are limited. CLINICALTRIAL PROSPERO CRD42024533180; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=533180

Point cloud registration is used to create a complete environment map with multiple pieces of scans. It is used mostly to guide autonomous robots when the surrounding is complicated. However, the problem is hard to solve so deep learning networks are often utilized. Most neural networks proposed need manual annotation and selection of patches which is time-consuming and makes the model only work for one particular scenario. In this work, a new method is studied in which a self-supervised learning technique is incorporated into the network. The input to this network is one raw point cloud without any labels or annotations, and as a result, the need for annotating or selecting patches is removed. A key point sampling process is also implemented in this network, which filters non-relevant points and further improves the performance. Based on our experiments, this self-supervised model has the capability of performing better than those requiring manual processing of the input.