Deep Learning Approach Predicts Longitudinal Retinal Nerve Fiber Layer Thickness Changes by Jalil Jalili, Evan Walker, Christopher Bowd, Akram Belghith, Michael H. Goldbaum, Massimo A. Fazio, Christopher A. Girkin, Carlos Gustavo De Moraes, Jeffrey M. Liebmann, Robert N. Weinreb, Linda M. Zangwill, Mark Christopher

“…We evaluated four models: linear regression (LR), support vector regression (SVR), gradient boosting regression (GBR), and a custom 1D convolutional neural network (CNN). The GBR model achieved the best performance in predicting pointwise RNFL thickness changes (MAE = 5.2 μm, R² = 0.91), while the custom 1D CNN excelled in predicting changes to average global and sectoral RNFL thickness, providing greater resolution and outperforming the traditional models (MAEs from 2.0–4.2 μm, R² from 0.94–0.98). …”
Get full text

Multilevel Feature Cross-Fusion-Based High-Resolution Remote Sensing Wetland Landscape Classification and Landscape Pattern Evolution Analysis by Sijia Sun, Biao Wang, Zhenghao Jiang, Ziyan Li, Sheng Xu, Chengrong Pan, Jun Qin, Yanlan Wu, Peng Zhang

“…To address these issues, this study proposes the multilevel feature cross-fusion wetland landscape classification network (MFCFNet), which combines the global modeling capability of Swin Transformer with the local detail-capturing ability of convolutional neural networks (CNNs), facilitating discerning intraclass consistency and interclass differences. …”
Get full text

Unsupervised learning-based panoramic unfolded image stitching method for rock mass borehole wall by XIAO Yu, LI Zehao, WANG Chao

“…A global and local deformation offset calculation network module precisely aligned spatial features of the images. …”
Get full text

Fault diagnosis of ZDJ7 railway point machine based on improved DCNN and SVDD classification by Zengshu Shi, Yiman Du, Xinwen Yao

“…First, the depthwise separable convolution in the Xception structure is used to optimize the extraction of fault features. …”
Get full text

DSGAU: Dual-Scale Graph Attention U-Nets for Hyperspectral Image Classification With Limited Samples by Hongzhuang Ji, Leying Song, Zhaohui Xue, Hongjun Su

“…This enables the simultaneous learning of local spectral features and global contextual patterns within HSI data. However, the convolutional operations in traditional GCNs require the inclusion of all data points during graph construction, leading to significant computational overhead, particularly for large-scale datasets. …”
Get full text

Enhancing Small Language Models for Graph Tasks Through Graph Encoder Integration by Dongryul Oh, Sujin Kang, Heejin Kim, Dongsuk Oh

“…Graphs inherently encode intricate structural dependencies, requiring models to effectively capture both local and global relationships. …”
Get full text

Occlusion Removal in Light-Field Images Using CSPDarknet53 and Bidirectional Feature Pyramid Network: A Multi-Scale Fusion-Based Approach by Mostafa Farouk Senussi, Hyun-Soo Kang

“…To preserve efficiency without sacrificing the quality of the extracted feature, our model uses separable convolutional blocks. A simple refinement module based on half-instance initialization blocks is integrated to explore the local details and global structures. …”
Get full text

A Graphite Ore Grade Recognition Method Based on Improved Inception-ResNet-v2 Model by Xueyu Huang, Renjie Pan, Jionghui Wang

“…Key improvements include: 1) To enhance the extraction of global feature information from graphite mine data, a global average pooling branch is incorporated into the Inception-resnet architecture. 2) Incorporating a <inline-formula> <tex-math notation="LaTeX">$1\times 1$ </tex-math></inline-formula> convolutional layer at the tail of the model to control channel dimensions and employing the LeakyReLU activation function to address the limitations of the ReLU activation function. 3) Designing an LDP-Conv structure to replace certain <inline-formula> <tex-math notation="LaTeX">$3\times 3$ </tex-math></inline-formula> convolutions and incorporating a channel attention mechanism to improve feature capture. 4) Optimizing the Stem module to expand the early-stage receptive field and reconstructing the network architecture. …”
Get full text

Graph-Based Few-Shot Learning for Synthetic Aperture Radar Automatic Target Recognition with Alternating Direction Method of Multipliers by Jing Jin, Zitai Xu, Nairong Zheng, Feng Wang

“…To address this challenge, we propose a novel few-shot learning (FSL) framework: the alternating direction method of multipliers–graph convolutional network (ADMM-GCN) framework. ADMM-GCN integrates a GCN with ADMM to enhance SAR ATR under limited data conditions, effectively capturing both global and local structural information from SAR samples. …”
Get full text

Machine learning for predicting Plasmodium liver stage development in vitro using microscopy imaging by Corin F. Otesteanu, Reto Caldelari, Volker Heussler, Raphael Sznitman

Get full text

ToxDL 2.0: Protein toxicity prediction using a pretrained language model and graph neural networks by Lin Zhu, Yi Fang, Shuting Liu, Hong-Bin Shen, Wesley De Neve, Xiaoyong Pan

Get full text

A novel deep learning method for large-scale analysis of bone marrow adiposity using UK Biobank Dixon MRI data by David M. Morris, Chengjia Wang, Giorgos Papanastasiou, Calum D. Gray, Wei Xu, Samuel Sjöström, Sammy Badr, Julien Paccou, Scott IK Semple, Tom MacGillivray, William P. Cawthorn

Get full text

Breast Cancer Histopathological Image Classification Based on High-Order Modeling and Multi-Branch Receptive Fields by Mengda Zhao, Cunqiao Hou, Lu Cao, Jianxin Zhang

“…Additionally, HoRFNet integrates a matrix power normalization strategy in the covariance pooling module to model the global interactions between convolutional features, thereby improving the higher-order representation of complex textures and structural relationships in tissue images. …”
Get full text

Non-end-to-end adaptive graph learning for multi-scale temporal traffic flow prediction. by Kang Xu, Bin Pan, MingXin Zhang, Xuan Zhang, XiaoYu Hou, JingXian Yu, ZhiZhu Lu, Xiao Zeng, QingQing Jia

“…Furthermore, the graph structure is dynamically updated using a weighted summation approach.Experiments demonstrate that the proposed method significantly improves prediction accuracy on the PeMSD4 and PeMSD8 datasets. …”
Get full text

Application of Partial Differential Equation Image Classification Methods to the Aesthetic Evaluation of Images by Feifeng Liu, Weihu Wang

“…The structure of a convolution kernel learned by using parallel network structure achieves better classification performance. …”
Get full text

HMA-Net: a hybrid mixer framework with multihead attention for breast ultrasound image segmentation by Soumya Sara Koshy, L. Jani Anbarasi

“…The model achieved a Jaccard index of 98.04% and 94.84% and a Dice similarity coefficient of 99.01% and 97.35% on the BUSI and BrEaST datasets, respectively.DiscussionThe ConvMixer and ConvNeXT modules are integrated with convolution-enhanced multihead attention, which enhances the model's ability to capture local and global contextual information. …”
Get full text

DSS-MobileNetV3: An Efficient Dynamic-State-Space- Enhanced Network for Concrete Crack Segmentation by Haibo Li, Yong Cheng, Qian Zhang, Lingkun Chen

“…The DSS-MobileNetV3 adopts a U-shaped encoder–decoder architecture, and a dynamic-state-space (DSS) block is designed into the encoder to improve the MobileNetV3 bottleneck module in modeling global dependencies. The DSS block improves the MobileNetV3 model in structural perception and global dependency modeling for complex crack morphologies by integrating dynamic snake convolution and a state space model. …”
Get full text

WDM-UNet: A Wavelet-Deformable Gated Fusion Network for Multi-Scale Retinal Vessel Segmentation by Xinlong Li, Hang Zhou

“…To address these limitations, we propose WDM-UNet, a novel spatial-wavelet dual-domain fusion architecture that integrates spatial and wavelet-domain representations to simultaneously enhance the local detail and global context. The encoder combines a Deformable Convolution Encoder (DCE), which adaptively models complex vascular structures through dynamic receptive fields, and a Wavelet Convolution Encoder (WCE), which captures the semantic and structural contexts through low-frequency components and hierarchical wavelet convolution. …”
Get full text

A novel pansharpening method based on cross stage partial network and transformer by Yingxia Chen, Huiqi Liu, Faming Fang

“…Abstract In remote sensing image fusion, the conventional Convolutional Neural Networks (CNNs) extract local features of the image through layered convolution, which is limited by the receptive field and struggles to capture global features. …”
Get full text

Rotten strawberry classification based on EfficientNet V2 algorithm fused with GCN and CA-Transformer by WANG Wei, YANG Shizhong, GONG Yucheng, GAO Sheng, DENG Zhaopeng

“…Secondly, this study integrated the Transformer structure with attention into the backbone of the baseline model, replacing some convolution operations with this structure to achieve the fusion of global and local features, thereby better identifying the rottenness of strawberries. …”
Get full text