A Multi-Scale Adaptive Fusion Network: End-to-End Interpretable Small-Sample Classifier for Motor Imagery EEG by Qiulei Han, Yan Sun, Ze Song, Hongbiao Ye, Tingwei Chen, Jian Zhao

“…Gradient-weighted Class Activation Mapping generated EEG topography maps visualized the brain regions of interest of the model, and the experimental results show that the decision-making process of the model in the MI-EEG decoding task is in accordance with neurophysiological principles. …”
Get full text

Experimental Study on Mixed Combustion Characteristics of Methanol/Diesel Pool Fires in Engine Rooms of Hybrid Ships by Jiaqi Dong, Zhongzheng Wu, Jinqi Han, Jianghao Li, Jiacheng Liu, Yunfeng Yan, Liang Wang

“…Therefore, a modification was made to the classical flame height model to account for these effects. Moreover, a prediction model for the burning rate of methanol/diesel pool fires was established based on transient temperature variations within the fuel layer. …”
Get full text

Formation of cheese pattern when using monospecies cultures by G. M. Sviridenko, O. M. Shukhalova, D. S. Vakhrusheva, D. S. Mamykin

“…It has been established that the use of a culture of citrate-fermenting lactococci L. diacetylactis ensures a stable lactic acid fermentation process during the production and ripening of the model and control cheeses, the formation of an elastic consistency and the desired pattern with eyes of a regular round shape. …”
Get full text

Deep blur detection network with boundary-aware multi-scale features by Xiaoli Sun, Qiwei Wang, Xiujun Zhang, Chen Xu, Weiqiang Zhang

“…First, the VGG-16 network is used to extract the deep features from multi-scale layers. Contrast layers and deconvolutional layers are added to make the difference between the blurred areas and clear areas more prominent. …”
Get full text

A Ship Detector Design Based on Deep Convolutional Neural Networks for Satellite Images by Ferhat Ucar, Deniz Korkmaz

“…Visual and graphicalresults of features at various layers and deconvolutions are also demonstratedfor a better understanding of the basic process.…”
Get full text

Research on Online Monitoring of Partial Discharge of Insulation Defects in Air Switchgear Based on Characteristic Gases by Yi Tian, Haotian Niu, Shuai Wang, Guixin Zhu

“…After that, the validity of the numerical model to describe the partial discharge process of air switchgear insulation defects is verified by the volt-ampere characteristic curve. …”
Get full text

The Study of Roadside Visual Perception in Internet of Vehicles Based on Improved YOLOv5 and CombineSORT by LI Xiaohui, YANG Jie, XIA Qin

“…In this process, the loss is calculated based on the differences in length, width, and diagonal between the detection and ground-truth boxes, and batch normalization (BN) layer sparsification is applied for convolutional channel filtering. …”
Get full text

Cross-Domain Transfer Learning Architecture for Microcalcification Cluster Detection Using the MEXBreast Multiresolution Mammography Dataset by Ricardo Salvador Luna Lozoya, Humberto de Jesús Ochoa Domínguez, Juan Humberto Sossa Azuela, Vianey Guadalupe Cruz Sánchez, Osslan Osiris Vergara Villegas, Karina Núñez Barragán

“…We propose a CNN trained on digital mammograms with three common resolutions: 50, 70, and 100 <inline-formula><math display="inline"><semantics><mi mathvariant="sans-serif">μ</mi></semantics></math></inline-formula>m. The architecture processes individual 1 cm² patches extracted from the mammograms as input samples and includes a MobileNetV2 backbone, followed by a flattening layer, a dense layer, and a sigmoid activation function. …”
Get full text

The manufacture of AAV for gene therapy applications using a closed, semi-automated hollow-fiber bioreactor by Adrien Soula, Florian Leseigneur, Amna Anwar, Bilal Ozdoganoglu, Jagan Gurung, Hamza Bhatti, Juline Guenat, Quentin Bazot, Majahar Sayed, Carolina Pinto Ricardo, Lily Li, Katerina Farukshina, Tony Bou Kheir, Hadi Mirmalek-Sani, Gregory Berger, Julie Kerby, Jonathan Appleby, Michael Delahaye

“…A cost of goods model comparing the Quantum bioreactor with the current standard HYPERStack36 and Corning CellSTACK 10-layer systems demonstrated that the Quantum bioreactor reduced the number of open steps by more than 40-fold, production time by up to 3.6-fold (HYPERStack36) and 7.5-fold (CellSTACK 10-layer), and costs by up to 2-fold (HYPERStack36) and 20.7-fold (CellSTACK 10-layer). …”
Get full text

Landfast ice in the Kara Sea stabilizes the Arctic halocline and may slow down Atlantification of the Eurasian Basin by Yuqing Liu, Martin Losch, L. Bruno Tremblay, Markus Janout

“…Abstract Observations show an Atlantification of the Eurasian Basin of the Arctic Ocean, with deeper penetration, shoaling, and ventilation of Atlantic waters in the eastern Arctic and an associated weakening of the cold halocline layer. These processes have a profound impact on the sea ice cover above and potentially on the transition of the Arctic to a seasonal ice cover. …”
Get full text

Leveraging Attribute Interaction and Self-Training for Graph Alignment via Optimal Transport by Songyang Chen, Youfang Lin, Ziyuan Zeng, Mengyang Xue

“…First, we enable the interaction of different dimensions of node attributes in the Gromov–Wasserstein learning process, while simultaneously integrating multi-layer graph structural information and node embeddings into the design of the intra-graph cost, which yields more expressive power with theoretical guarantee. …”
Get full text

Cold Expansion Strengthening of 7050 Aluminum Alloy Hole: Structure, Residual Stress, and Fatigue Life by Qi Li, Qichao Xue, Quansheng Hu, Ting Song, Yonghui Wang, Siyang Li

“…The research results indicate that the cold expansion process of slotted bushing holes can enhance the residual stress distribution around the hole and form a strong residual stress layer. …”
Get full text

Kinetics of extraction from water solutions of Fe³⁺, Cu²⁺ and Pb²⁺ + ions by filter loading, obtained by blinding plasters by A. I. Teran

“…Three models that best correspond to real processes are selected. …”
Get full text

Deep learning for real-time P-wave detection: A case study in Indonesia's earthquake early warning system by Adi Wibowo, Leni Sophia Heliani, Cecep Pratama, David Prambudi Sahara, Sri Widiyantoro, Dadan Ramdani, Mizan Bustanul Fuady Bisri, Ajat Sudrajat, Sidik Tri Wibowo, Satriawan Rasyid Purnama

“…We have integrated several post-processing techniques to further bolster the model's robustness against noise. …”
Get full text

Efficient adsorptive removal of hazardous congo red dye using Ce-BTC@microcrystalline cellulose composite by Mostafa A. Sayed, Reda M. Abdelhameed, Ibrahim H. A. Badr, Ali M. Abdel-Aziz

“…The adsorption kinetics followed a pseudo-second-order model (R2 = 0.988), and both intraparticle and boundary layer diffusion influenced the rate-limiting step of the adsorption process. …”
Get full text

Plastic to apparel: an analysis of sustainable purchasing intention using a machine learning ensemble by Carmella Andrea L. Cabrera, Ardvin Kester S. Ong, John Francis T. Diaz, Maela Madel L. Cahigas, Ma. Janice J. Gumasing

“…Data preprocessing using feature selection and correlation analysis was conducted to validate the available data, performed data cleaning process, and data aggregation. Several iterative processes were employed to generate the optimum classification model—obtaining a 92% accuracy for RFC and 91% for ANN at 150 epochs under 30 hidden layer nodes. …”
Get full text

Experimental Investigation on Mechanical Behaviour of 3D-Printed Part for Brake Pedal Applications by T. Niranjan, B. Singaravel, S. Srinivasulu Raju, R. Thanigaivelan

“…It also showed that Infill density (31.06%), material type (21.54%) and Layer Thickness (20.41%) are the most influencing process parameters of Impact strength. …”
Get full text

The Mechanism of Introduction of Iron and Copper Atoms into the Interlayer Space of Ir-Pyrolyzed Acrylonitrile Nanopolymer by I.V. Zaporotskova, N.A. Anikeev, L.V. Kozitov

“…The process of introducing metal atoms occurs through the open end of a two-layer PPAN and modeled step by step approach (step 0,1 Å) to the fictitious atom, which is located in the center of the cluster between the layers.…”
Get full text

Multichannel convolutional transformer for detecting mental disorders using electroancephalogrpahy records by Mamadou Dia, Ghazaleh Khodabandelou, Syed Muhammad Anwar, Alice Othmani

“…Before feeding the model as low-level features, the input is pre-processed using a common spatial pattern filter, a signal space projection filter, and a wavelet denoising filter. …”
Get full text

Energy-Efficient Deep Learning for Cloud Detection Onboard Nanosatellite by Imane Khalil, Amina Daghouri, Mohammed Alae Chanoui, Zouhair Guennoun, Adnane Addaim

“…This work, conducted within the project of the UM5-EOSat nanosatellite mission, utilized captured images from the Gecko imager for model evaluation. The customized SegNet architecture, tailored with minimal kernels and layers, achieved an accuracy of 93.50&#x0025;, effectively balancing performance and computational efficiency. …”
Get full text