The use of very-high-resolution images to extract urban, suburban and rural roads has important application value. However, it is still a problem to effectively extract the road area occluded by roadside tree canopy or high-rise buildings to maintain the integrity of the extracted road area, the smoothness of the sideline and the connectivity of the road network. This paper proposes an innovative Cascaded Attention DenseUNet (CADUNet) semantic segmentation model by embedding two attention modules, such as global attention and core attention modules, in the DenseUNet framework. First, a set of cascaded global attention modules are introduced to obtain the contextual information of the road; secondly, a set of cascaded core attention modules are embedded to ensure that the road information is transmitted to the greatest extent among the dense blocks in the network, and further assist the global attention module in acquiring multi-scale road information, thereby improving the connectivity of the road network while restoring the integrity of the road area shaded by the tree canopy and high-rise buildings. Based on binary cross entropy, an adaptive loss function is proposed for network parameter tuning. Experiments on the Massachusetts road dataset and the DeepGlobe-CVPR 2018 road dataset show that this semantic segmentation model can effectively extract the road area shaded by tree canopy and improve the connectivity of the road network.

Due to the superiority of convolutional neural networks, many deep learning methods have been used in image classification. The enormous difference between natural images and remote sensing images makes it difficult to directly utilize or modify existing CNN models for remote sensing scene classification tasks. In this article, a new paradigm is proposed that can automatically design a suitable CNN architecture for scene classification. A more efficient search framework, RS-DARTS, is adopted to find the optimal network architecture. This framework has two phases. In the search phase, some new strategies are presented, making the calculation process smoother, and better distinguishing the optimal and other operations. In addition, we added noise to suppress skip connections in order to close the gap between trained and validation processing and ensure classification accuracy. Moreover, a small part of the neural network is sampled to reduce the redundancy in exploring the network space and speed up the search processing. In the evaluation phase, the optimal cell architecture is stacked to construct the final network. Extensive experiments demonstrated the validity of the search strategy and the impressive classification performance of RS-DARTS on four public benchmark datasets. The proposed method showed more effectiveness than the manually designed CNN model and other methods of neural architecture search. Especially, in terms of search cost, RS-DARTS consumed less time than other NAS methods.

Enhancers are located upstream or downstream of key deoxyribonucleic acid (DNA) sequences in genes and can adjust the transcription activity of neighboring genes. Identifying enhancers and determining their functions are important for understanding gene regulatory networks and expression regulatory mechanisms. However, traditional enhancer recognition relies on manual feature engineering, which is time-consuming and labor-intensive, making it difficult to perform large-scale recognition analysis. In addition, if the original dataset is too small, there is a risk of overfitting. In recent years, emerging methods, such as deep learning, have provided new insights for enhancing identification. However, these methods also present certain challenges. Deep learning models typically require a large amount of high-quality data, and data acquisition demands considerable time and resources. To address these challenges, in this paper, we propose a data-augmentation method based on generative adversarial networks to solve the problem of small datasets. Moreover, we used regularization methods such as weight decay to improve the generalizability of the model and alleviate overfitting. The Transformer encoder was used as the main component to capture the complex relationships and dependencies in enhancer sequences. The encoding layer was designed based on the principle of k-mers to preserve more information from the original DNA sequence. Compared with existing methods, the proposed approach made significant progress in enhancing the accuracy and strength of enhancer identification and prediction, demonstrating the effectiveness of the proposed method. This paper provides valuable insights for enhancer analysis and is of great significance for understanding gene regulatory mechanisms and studying disease correlations.

Introduction: Depression is a rising global health issue, particularly among adolescents, with university students facing distinct mental health challenges.Methods: This proof-of-concept study explores smartphone sensor-based depression detection in Chinese university campus settings using a small sample of 12 participants. We utilized data from accelerometers, gyroscopes, and light sensors to establish associations between smartphone-derived behavioral patterns and PHQ-9 scores, a standard depression measure. A customized data processing scheme tailored to campus life enabled the extraction of 18 feature sequences reflecting depressive symptoms. Feature selection was conducted using Pearson correlation, and model validation was performed using leave-one-out cross-validation with common classification algorithms.Results: The results yielded accuracy rates between 73.11% and 88.24%. Findings showed negative correlations between PHQ-9 scores and dietary regularity, bedtime, and physical activity levels.Discussion: This pioneering study highlights smartphone sensors' potential for early depression detection in Chinese higher education, supporting non-invasive mental health interventions.