Edge detection methods are significant in medical imaging-assisted diagnosis. However, existing methods based on grayscale gradient computation still need to be optimized in practicality, especially in terms of actual visual quality and sensitivity to image contrast. To optimize the visualization and enhance the robustness of contrast changes, we propose the Contrast Invariant Edge Detection (CIED) method. CIED combines Gaussian filtering and morphological processing methods to preprocess medical images. It utilizes the three Most Significant Bit (MSB) planes and binary images to detect and extract significant edge information. Each bit plane is used to detect edges in 3 × 3 blocks by the proposed algorithm, and then the edge information from each plane is fused to obtain an edge image. This method is generalized to common types of images. Since CIED is based on binary bit planes and eliminates complex pixel operations, it is faster and more efficient. In addition, CIED is insensitive to changes in image contrast, making it more flexible in its application. To comprehensively evaluate the performance of CIED, we develop a medical image dataset and conduct edge image and contrast evaluation experiments based on these images. The results show that the average precision of CIED is 0.408, the average recall is 0.917, and the average F1-score is 0.550. The results indicate that CIED is not only more practical in terms of visual effects but also robust in terms of contrast invariance. The comparison results with other methods also confirm the advantages of CIED. This study provides a novel approach for edge detection within medical images.

In recent years, the number of telecom frauds has increased significantly, causing substantial losses to people’s daily lives. With technological advancements, telecom fraud methods have also become more sophisticated, making fraudsters harder to detect as they often imitate normal users and exhibit highly similar features. Traditional graph neural network (GNN) methods aggregate the features of neighboring nodes, which makes it difficult to distinguish between fraudsters and normal users when their features are highly similar. To address this issue, we proposed a spatio-temporal graph attention network (GDFGAT) with feature difference-based weight updates. We conducted comprehensive experiments on our method on a real telecom fraud dataset. Our method obtained an accuracy of 93.28%, f1 score of 92.08%, precision rate of 93.51%, recall rate of 90.97%, and AUC value of 94.53%. The results showed that our method (GDFGAT) is better than the classical method, the latest methods and the baseline model in many metrics; each metric improved by nearly 2%. In addition, we also conducted experiments on the imbalanced datasets: Amazon and YelpChi. The results showed that our model GDFGAT performed better than the baseline model in some metrics.

The integration of virtual museums into education has emerged as an innovative approach embraced by both teachers and learners, reflecting the broader impact of virtual reality (VR) applications in education. This study puts forward a pedagogical framework for utilizing virtual museums in teaching art history and investigating their impact on the art history curriculum. In this context, two free online museums are used as teaching materials, representing 3D interactive learning environments that enable immersive exploration of cultural heritage. Grounded in the Theory of Technology-Mediated Learning, this research adopts a hybrid methodological approach to track the art history courses of 75 Chinese undergraduates through experiments, questionnaires, and structured interviews over a four-week period. The findings demonstrate that virtual museum-integrated instruction significantly enhances learning effectiveness over sustained use, actively promotes learner engagement, and fosters greater autonomy. Importantly, learners prioritize educational value and authenticity in virtual museum features, while also expressing a strong preference for technologically mature platforms. This research contributes to understanding the impact of VR on digital transformation in the educational sector by providing a validated instructional model that integrates virtual museums into art history curricula, offering educators a replicable framework for implementation. Future studies should investigate the relationship between emotional engagement and academic performance within virtual museums to further refine both pedagogical strategies and educational virtual reality design.

Social media platforms increasingly function as informal education systems for programming learning, yet the systemic support structures these environments provide remain poorly understood. We analyzed 40,004 comments from programming tutorial videos on a major social media platform (2016–April 2025) to identify patterns of learner support needs at scale. Using BERTopic, we identified twelve discussion themes. We then consolidated these themes into a learner-needs typology based on their dominant support functions: instructional-oriented needs, operational support needs, and knowledge-constructionneeds. We mapped this typology onto the Community of Inquiry (CoI) framework to assess its explanatory coverage. This mapping revealed a critical systemic gap. Operational support needs, covering environment configuration, tool integration, dependency management, and technical troubleshooting, constituted the largest category (44.53% of theme-level discourse), exceeding both knowledge-construction needs (28.42%) and instructional-oriented needs (26.95%). Learners repeatedly described these infrastructure-level challenges as disrupting their attempts to engage with content, execute code for testing ideas, and coordinate with peers, yet these operational readiness needs are not fully specified by CoI’s traditional presences. Social presence did not emerge as a standalone theme at the topic-modeling level; rather, social cues were often embedded within task-oriented troubleshooting. Based on these findings, we propose Technical Presence as a context-sensitive extension to the CoI framework, defined as the extent to which a learning community enables operational readiness through accessible infrastructure support and collaborative troubleshooting. As an infrastructural support condition, Technical Presence supports operational readiness within tool-dependent, practice-based learning: when learners report infrastructure failure, the conditions for enacting instructional design, cognitive inquiry, and peer collaboration are correspondingly weakened. These findings carry implications for content creators, platform developers, and education system designers seeking to strengthen the infrastructural foundations of technology-enhanced learning at scale.