Ultra-Wide-Band and Polarization-Insensitive Metamaterial Absorber With Resistive Ink Sheet for RCS Reduction Applications by Ramesh Amugothu, Vakula Damera

“…The structure of the unit cell is designed using a dielectric FR4 substrate with resistive ink patterns printed on it. With the help of computational optimization using the Ansys HFSS tool, the proposed unit cell's various geometric parameters are fine-tuned to achieve outstanding results. …”
Get full text

Deep learning-based action recognition for joining and welding processes of dissimilar materials by Tao He, Xinyuan Jin, Yiming Zou

“…To address these limitations, we propose a novel deep learning-based framework specifically designed for action recognition in joining and welding tasks involving dissimilar materials.MethodsOur proposed model, the Multi-Scale Spatiotemporal Attention Network (MS-STAN), leverages advanced hierarchical feature extraction techniques and attention mechanisms to capture fine-grained spatiotemporal patterns across varying scales. …”
Get full text

P-GELU: A Novel Activation Function to Optimize Whisper for Darija Speech Translation by Maria Labied, Abdessamad Belangour, Mouad Banane

“…Furthermore, P-GELU reduces training loss, improves feature retention, and enhances linguistic adaptability, making it a more effective alternative for speech translation tasks involving phonetic variability, code-switching, and limited training data. …”
Get full text

Enhanced IoT-Based Face Mask Detection Framework Using Optimized Deep Learning Models: A Hybrid Approach With Adaptive Algorithms by Parul Dubey, Pushkar Dubey, Celestine Iwendi, Cresantus N. Biamba, Deepak Dasaratha Rao

“…In the present study, a hybrid Flame-Sailfish Optimization (HFSO)-based deep learning framework is proposed. It combines the feature extraction capabilities of ResNet50 with the efficiency of MobileNetV2. …”
Get full text

Rep-MobileViT: Texture and Color Classification of Solid Wood Floors Based on a Re-Parameterized CNN-Transformer Hybrid Model by Anning Duanmu, Sheng Xue, Zhenye Li, Yajun Zhang, Chao Ni

“…Specifically, the RepAIRB module is introduced, incorporating an asymmetric convolutional block (ACB) and a re-parameterized structure within the inverted residual block (IRB) module to enhance the network’s receptive field without increasing computational costs. Additionally, the RepMFF-MobileViT modu le is designed, utilizing a multi-feature fusion strategy and the RepTransformer structure to reduce the number of parameters while improving the network’s expressiveness and stability. …”
Get full text

Preoperative prediction of pulmonary ground-glass nodule infiltration status by CT-based radiomics combined with neural networks by Kun Mei, Zikang Feng, Hui Liu, Min Wang, Chao Ce, Shi Yin, Xiaoying Zhang, Bin Wang

“…Regions of interest (ROIs) within the CT lung window level were manually delineated using ITK-SNAP software, enabling the extraction of relevant CT imaging features, including morphological descriptors, first-order statistical parameters, texture attributes, and high-order characteristics. …”
Get full text

Lightweight deep neural network for contour detection and extraction of wheat spikes in complex field environments by Xin Xu, Haiyang Zhang, Jiangchuan Lu, Ziyi Guo, Juanjuan Zhang, Jibo Yue, Hongbo Qiao, Xinming Ma

“…The algorithm innovatively constructs a lightweight depthwise separable network (LDSNet) as backbone, balancing computational efficiency and accuracy through channel re-parameterization strategy; incorporates an Efficient Local Attention (ELA) module to build feature enhancement networks, and employs dual-path feature fusion mechanisms to strengthen edge texture responses, significantly improving discrimination of overlapping spikes and complex backgrounds. …”
Get full text

A turbulence reduced order model based on non-interpolated convolutional autoencoder by WU Pin, ZHANG Bo, SONG Chao, ZHOU Zhu

“…This paper introduces an innovative solution: a non-interpolated convolutional autoencoder, designed to extract nonlinear features from the flow field while curbing parameter count, evading interpolation errors, and mitigating additional computational burdens. …”
Get full text

Advanced holographic convolutional dense networks and Tangent runner optimization for enhanced polycystic ovarian disease classification by Prathibanandhi Jeyashanker, Annie Grace Vimala Georgewilliam Sundaram, Padmakala Sadagopan, Abid Yahya, Ravi Samikannu, Irfan Anjum Badruddin, Sarfaraz Kamangar, Mohamed Ghouse Shukur

Get full text

Lightweight multilayer interactive attention network for aspect-based sentiment analysis by Wenjun Zheng, Shunxiang Zhang, Cheng Yang, Peng Hu

“…Such correlation degree is calculated by multiple computational layers with neural attention models. Third, we use a parameter-sharing strategy among the computational layers. …”
Get full text

An improved framework to prioritize sub-watersheds: Construction of Watershed Morphometric Composite Index (WMCI) for mountain soil conservation by Shravan Kumar Ghimire, Yafeng Lu, Yukuan Wang

“…The study has introduced a feature scaling approach to the morphometric measurement based on the relationship of parameters with soil erodibility and constructed a composite index of parameters. …”
Get full text

Prediction of heart failure risk factors from retinal optical imaging via explainable machine learning by Sona M. Al Younis, Samit Kumar Ghosh, Hina Raja, Feryal A. Alskafi, Siamak Yousefi, Siamak Yousefi, Ahsan H. Khandoker

Get full text

Brosimine B and the biphasic dose-response: insights into hormesis and retinal neuroprotection by Susanne Suely Santos Fonseca, Susanne Suely Santos Fonseca, Natacha M. S. Port’s, Gisele Priscila Soares Aguiar, Eliã P. Botelho, Nádia M. G. Couto, Wandson Braamcamp Souza Pinheiro, André Salim Khayat, Elizabeth S. Yamada, Edmar T. Costa, Chubert Bernardo C. Sena, Mara Silvia P. Arruda, Carlomagno P. Bahia, Antonio Pereira

“…However, higher concentrations (>10 µM) induced cytotoxic effects. A computational modeling approach using a hormetic (inverted U-shaped) model revealed biologically interpretable parameters, including a peak response at 10.2 µM and a hormetic zone width (σ = 6.5 µM) (R2 = 0.984).DiscussionThese results confirm that Brosimine B exhibits hormetic neuroprotective effects within a well-defined concentration window, supporting its potential as a therapeutic agent for oxidative stress–related retinal damage. …”
Get full text

Hybrid Reinforcement Learning-Based Collision Avoidance Algorithm for Autonomous Vehicle Clusters by Chubing Guo, Jianshe Wu, Panzheng Luo, Zhigang Wang, Kai Zhang, Ziyi Yang, Zengfa Dou, Kan Song

Get full text

Explainable post hoc portfolio management financial policy of a Deep Reinforcement Learning agent. by Alejandra de-la-Rica-Escudero, Eduardo C Garrido-Merchán, María Coronado-Vaca

“…We empirically illustrate it by successfully identifying key features influencing investment decisions, which demonstrate the ability to explain the agent actions in prediction time. …”
Get full text

1.7 MV Tandem Accelerator Beam Tuning Optimization by JIANG Ran, LI Ailing, CUI Baoqun, TANG Bing, CHEN Haonan, WANG Yunfeng

“…This interface empowered efficient upper-level control and real-time monitoring functions, thereby significantly bolstering the usability and practicality of the system. The CSS interface features a user-friendly graphical user interface (GUI) that allows operators to monitor and adjust parameters in real-time with ease, enhancing the overall user experience and operational efficiency. …”
Get full text

ES-UNet: efficient 3D medical image segmentation with enhanced skip connections in 3D UNet by Minyoung Park, Seungtaek Oh, Junyoung Park, Taikyeong Jeong, Sungwook Yu

“…Methods We propose ES-UNet, a 3D segmentation architecture that achieves superior segmentation performance while offering competitive efficiency across multiple computational metrics, including memory usage, inference time, and parameter count. …”
Get full text

Adaptive Global Dense Nested Reasoning Network into Small Target Detection in Large-Scale Hyperspectral Remote Sensing Image by Siyu Zhan, Yuxuan Yang, Muge Zhong, Guoming Lu, Xinyu Zhou

“…Secondly, we present a novel approach for computing physicochemical parameters, which enhances the spectral characteristics of targets while minimizing environmental interference. …”
Get full text

Pork-YOLO: Automated collection of pork quality traits by Jiacheng Wei, Xi Tang, Jinxiu Liu, Ting Luo, Yan Wu, Junhui Duan, Shijun Xiao, Zhiyan Zhang

“…Next, the StarNet backbone was integrated, employing star-shaped operations for high-dimensional feature representation while maintaining computational efficiency, along with a hierarchical convolutional structure to boost performance. …”
Get full text

Precise Spatial Prediction of Rice Seedlings From Large-Scale Airborne Remote Sensing Data Using Optimized Li-YOLOv9 by Jayakrishnan Anandakrishnan, Arun Kumar Sangaiah, Hendri Darmawan, Nguyen Khanh Son, Yi-Bing Lin, Mohammed J. F. Alenazi

“…The compact Li-YOLOv9 with approximately 9 million parameters is significantly lighter than the original YOLOv9 with 60 million parameters, making it ideal for resource-efficient onboard intelligence. …”
Get full text