Progressive multi-scale attention neural network for pneumonia classification in chest X-rays by Mohammad Reza Mahdiani

“…Unlike previous methods that overlook fine-grained edge information or fail to integrate multi-scale contextual features, our approach synergistically combines convolutional multi-scale feature extraction using depthwise separable convolutions with cross-layer feature fusion, Transformer blocks, advanced attention mechanisms, and a custom loss function that emphasizes diagnostically relevant edge details using Canny edge detection. …”
Get full text

Lightweight Deep Learning Model for Fire Classification in Tunnels by Shakhnoza Muksimova, Sabina Umirzakova, Jushkin Baltayev, Young-Im Cho

“…This model integrates MobileNetV3 for spatial feature extraction, Temporal Convolutional Networks (TCNs) for temporal sequence analysis, and advanced attention mechanisms, including Convolutional Block Attention Modules (CBAMs) and Squeeze-and-Excitation (SE) blocks, to prioritize critical features such as flames and smoke patterns while suppressing irrelevant noise. …”
Get full text

Kolmogorov–Arnold–Enhanced Nonlinear Expansions for Fine-Grained Feature Amplification in Robust Near-Shore SAR Vessel Discrimination by Hankang Wang, Gaopeng Huang, Zhao Huang, Xingru Huang, Zhaoyang Xu, Zhiwen Zheng, Shanwei Liu, Shiqing Wei, Jin Liu, Xiaoshuai Zhang

“…This structure preserves small-target details by propagating enriched semantic information across detection layers, mitigating feature loss during downsampling. …”
Get full text

DeepLASD countermeasure for logical access audio spoofing by Hamed Al-Tairi, Ali Javed, Tasawer Khan, Abdul Khader Jilani Saudagar

“…The model incorporates a SincConv layer for interpretable spectral processing, along with residual convolutional blocks that integrate attention for improved feature extraction. …”
Get full text

Multi-Class Brain Lesion Classification Using Deep Transfer Learning With MobileNetV3 by Ahmed Firas Majeed, Pedram Salehpour, Leili Farzinvash, Saeid Pashazadeh

Get full text

DEFORMATION WAVES AS A TRIGGER MECHANISM OF SEISMIC ACTIVITY IN SEISMIC ZONES OF THE CONTINENTAL LITHOSPHERE by S. I. Sherman

“…The method can be applied to estimate a dominating direction of movement of the epicentres, which corresponds to the phase velocity of the deformation wave disturbing meta-stability of the fault-block medium, leading to displacement of neighbouring blocks and thus causing a seismic event (Fig. 14). …”
Get full text

A Three-Stage-Concatenated Non-Linear MMSE Interference Rejection Combining Aided MIMO-OFDM Receiver and its EXIT-Chart Analysis by Jue Chen, Siyao Lu, Tsang-Yi Wang, Jwo-Yuh Wu, Chih-Peng Li, Soon Xin Ng, Robert G. Maunder, Lajos Hanzo

“…This reduced complexity NL detection algorithm is particularly well suited for practical hardware implementation using parallel processing at a low latency. …”
Get full text

WaveConv-sLSTM-KET: A Novel Framework for the Multi-Task Analysis of Oil Spill Fluorescence Spectra by Shubo Zhang, Menghan Li, Jing Li

“…By combining a Wavelet Transform CNN block, a scalar LSTM block, and a Kolmogorov–Arnold Network-Enhanced Transformer block, the framework enables simultaneous oil-type identification and thickness prediction without preprocessing or fully connected layers. …”
Get full text

Research on shale lamination types and logging characterization methods: A case study of the Funing Formation Member 2 in Gaoyou Sag, Subei Basin by TANG Lei, LIAO Wenting, XIA Lianjun, MA Jie, ZHANG Juan

“…The study identified intervals Ⅳ-3 to Ⅳ-7 and Ⅴ-6 to Ⅴ-8 in Funing Formation Member 2 as having well-developed laminations and higher total organic carbon (TOC) compared to other intervals, marking them as vertical shale oil sweet spot layers. The image edge detection method using electrical imaging well logging offers high accuracy for shale bedding identification and is suitable for detailed geological evaluation of vertical shale oil sweet spot layers in different blocks. …”
Get full text

Next-generation carbon nanotube electrochemical sensors for liquid biopsy applications by Samuel Rantataro, Eero Gustafsson, Ilkka Varjos

“…Contrasting that, we report the development of a highly-sensitive electrochemical sensor assay, which is based on pristine carbon nanotube (CNT) electrodes functionalized with casein blocking layer and sequence-specific probes for ctDNA. …”
Get full text

Modeling and Simulation of a Resonant-Cavity-Enhanced InGaAs/GaAs Quantum Dot Photodetector by W. W. Wang, F. M. Guo, Y. Q. Li

“…Our simulation results also showed that when an AlAs layer is inserted into the device structure as a blocking layer, ultralow dark current can be achieved, with dark current densities 0.0034 A/cm at 0 V and 0.026 A/cm at a reverse bias of 2 V. …”
Get full text

Imaging Monitoring Technology of Fractures Around Horizontal Wells with Multi-Stage Fracturing in Water Injection Development by ZHENG Gang, ZHU Hongzheng, LI Dajian, WANG Xu, TANG Xiangdong, SU Zubo

“…The imaging monitoring technology of the fractures around the well realizes the accurate identification of the water interval and the main interval of the remaining oil exploitation and has solved the problems of unclear out-of-pipe and in-layer flow, as well as the difficulty of identifying the direction for subsequent water blocking and oil recovery.…”
Get full text

Piezo1 Promotes Odontogenic Differentiation of Dental Pulp Stem Cells Under Stress Conditions by Xiaxia Wang, Shaojie Dong, Qianqian Dong, Xuefei Sun

“…Results: The Piezo1 protein was positively expressed in human dental pulp samples, especially in the odontoblast layer. Increased Piezo1 expression was also detected after odontogenic differentiation of DPSCs in vitro. …”
Get full text

Leveraging federated learning and edge computing for pandemic-resilient healthcare by Atlanta Choudhury, Kandarpa Kumar Sarma, Vyacheslav Gulvanskii, Dmitrii Kaplun, Lachit Dutta

“…In particular, the integration of the YoloV4 and SENET attention layer as part of a FL framework delivers dependable performance while addressing facemask detection (94.6%), incorrect facemask detection (98%), facemask classification (95.4%), social distance (96.1%), contact tracing (95.2%) and cyber attack detection (94.2%) while performing tasks like correct and incorrect, proper and improper facemask wearing, monitoring social distancing norms observance, and contact tracing.…”
Get full text

Encrypted traffic classification method based on convolutional neural network by Rongna XIE, Zhuhong MA, Zongyu LI, Ye TIAN

“…Aiming at the problems of low accuracy, weak generality, and easy privacy violation of traditional encrypted network traffic classification methods, an encrypted traffic classification method based on convolutional neural network was proposed, which avoided relying on original traffic data and prevented overfitting of specific byte structure of the application.According to the data packet size and arrival time information of network traffic, a method to convert the original traffic into a two-dimensional picture was designed.Each cell in the histogram represented the number of packets with corresponding size that arrive at the corresponding time interval, avoiding reliance on packet payloads and privacy violations.The LeNet-5 convolutional neural network model was optimized to improve the classification accuracy.The inception module was embedded for multi-dimensional feature extraction and feature fusion.And the 1*1 convolution was used to control the feature dimension of the output.Besides, the average pooling layer and the convolutional layer were used to replace the fully connected layer to increase the calculation speed and avoid overfitting.The sliding window method was used in the object detection task, and each network unidirectional flow was divided into equal-sized blocks, ensuring that the blocks in the training set and the blocks in the test set in a single session do not overlap and expanding the dataset samples.The classification experiment results on the ISCX dataset show that for the application traffic classification task, the average accuracy rate reaches more than 95%.The comparative experimental results show that the traditional classification method has a significant decrease in accuracy or even fails when the types of training set and test set are different.However, the accuracy rate of the proposed method still reaches 89.2%, which proves that the method is universally suitable for encrypted traffic and non-encrypted traffic.All experiments are based on imbalanced datasets, and the experimental results may be further improved if balanced processing is performed.…”
Get full text

The role of multimodal imaging in the diagnostics of non-compaction cardiomyopathy (a clinical case) by M. Yu. Kolesnyk, S. V. Fedkiv, M. V. Sokolova, O. V. Nikitiuk

“…It was described a clinical case of a 53-year old female with intact coronary arteries, who had been followed up for ten years with a diagnosis of dilated cardiomyopathy after the detection of a left bundle branch block. When performing standard echocardiography, a two-layer structure of the myocardium was found in the apical lateral segments of the left ventricle (LV) with the presence of hypertrabeculated internal layer and diffuse decrease in global LV contractility (three-dimensional ejection fraction 34 %). …”
Get full text

A Few-Shot SE-Relation Net-Based Electronic Nose for Discriminating COPD by Zhuoheng Xie, Yao Tian, Pengfei Jia

“…The model integrates residual blocks, BiGRU layers, and squeeze–excitation attention mechanisms to enhance feature-extraction efficiency. …”
Get full text

Development of White Cabbage, Coffee, and Red Onion Extracts as Natural Phosphodiesterase-4B (PDE4B) Inhibitors for Cognitive Dysfunction: In Vitro and In Silico Studies by Nazir Ahmad, Kaisun Nesa Lesa, Navista Sri Octa Ujiantari, Ari Sudarmanto, Nanang Fakhrudin, Zullies Ikawati

“…Thus, the combination extracts are a promising cognitive enhancer by blocking PDE4B activity.…”
Get full text

Cross-Scale Feature Blending Model for Surface Defect Identification in Machine Tool Elements Resilient to Contaminant Interference by Dong Wu, Chunhua Guo, Renpu Li, Zhigang Ma

“…The CSFB architecture innovates a distinct mechanism to feature fusion, where each layer receives direct input from all previous layers to blend entire-scale features up to the top layer, addressing discrepancies between features visible at multiple scales. …”
Get full text

Enhanced Transformer Network With High-Dimensional Attention Mechanism for Diabetic Retinopathy Classification by M. Rizvana, Sathiya Narayanan

“…In addition, the classification head present in the ViT block incorporates sequential dropout layer in addition to the standard linear dropout layer. …”
Get full text