NSTL国家科技图书文献中心

1. Graph neural network for fraud detection via context encoding and adaptive aggregation NSTL国家科技图书文献中心

Chaoli Lou | Yueyang Wang ... - 《Expert Systems with Application》 - 2025,261(Feb.) - 125473.1～125473.12 - 共12页

摘要： networks to fraud detection* still faces challenges of* | capturing two types of fraudulent behaviors: the context | encoding module which encodes the time intervals of out | address the challenges, we present Context-encoding-and | GNN-based fraud detection* model. To capture the*

关键词： Camouflage | Fraud detection | Graph neural network

2. An improved supervised contrastive learning with denoising diffusion probabilistic model for fault detection in industrial processes NSTL国家科技图书文献中心

Li, Daye | Dong, Jie ... - 《Transactions of The Institution of Chemical Engineers,Part B.Process Safety and Environmental Protection》 - 2025,194 - 350～359 - 共10页

摘要： constructed to filter high-quality out* of distribution* | The distribution of actual industrial process | data follows a Gaussian distribution. In this context | , we propose a novel fault detection* method based on* | improved supervised contrastive learning detection* model*

关键词： Fault detection | Denoising diffusion probabilistic model | Supervised contrastive learning | Out-of-distribution data | Process safety

3. A new method for tuning the CNN pre-trained models as a feature extractor for malware detection NSTL国家科技图书文献中心

Halit Bakir - 《Pattern analysis and applications》 - 2025,28(1) - 26.1～26.19 - 共19页

摘要： enhancing detection* performance. Using a dataset of 3000* | effectiveness of CNN-based malware detection. This work stands | out by focusing on the dynamic tuning of deep | malware detection, current approaches face notable | , achieving high detection* accuracy, and maintaining*

关键词： Android | Malware detection | Hyperparameters tuning | Convolutional neural network | Pre-trained model

4. Echocardiographic assessment of left atrial appendage morphology and function-an expert proposal by the German Working Group of Cardiovascular Ultrasound NSTL国家科技图书文献中心

Hagendorff, Andreas | Stoebe, Stephan ... - 《Clinical research in cardiology :》 - 2025,114(1) - 25～40 - 共16页

摘要： the left atrial appendage. The detection* of thrombi* | complications-especially in the context* of rhythm control. The* | morphology. This structure is a preferred region of | of the atrial wall. Blood stasis occurs primarily | in low flow conditions. One of the tasks of

关键词： Atrial fibrillation | Echocardiography | Left atrial appendage | LAA morphology | LAA function | LAA intervention

5. Behavioural ecology and developmental plasticity in Indian anuran tadpoles NSTL国家科技图书文献中心

Saidapur, Srinivas K... - 《Behavioral Ecology and Sociobiology》 - 2025,79(3) - 共20页

摘要：Studies on behavioural ecology of tadpoles of | morphological features. 2) The sensory basis of food or | predator detection* in herbivorous tadpoles is* | visual and chemical cues in prey detection/ capture. 3 | (kairomones, alarm cues or dietary metabolites of

关键词： Antipredator defences | Developmental plasticity | Kin recognition | Microhabitat choice | Predator-prey interactions | Tadpoles

6. Dual-Dynamic Cross-Modal Interaction Network for Multimodal Remote Sensing Object Detection NSTL国家科技图书文献中心

Wei Bao | Meiyu Huang ... - 《IEEE Transactions on Geoscience and Remote Sensing》 - 2025,63 - 1～13 - 共13页

摘要：Multimodal remote sensing object detection* (MM* | design of channel-gated spatial cross-attention (CSCA | -modal context* interaction through static convolution* | kernels to filter out* irrelevant information between* | -specific global context* to fuse features, achieving*

关键词： Feature extraction | Remote sensing | Object detection | Termination of employment | Redundancy | Optical imaging | Optical filters | Filters | Fuses | Uncertainty

7. Multi-modal Multitask Learning Model for Simultaneous Classification of Two Epilepsy Biomarkers NSTL国家科技图书文献中心

Nawara Mahmood Broti | Masaki Sawada ... - 《Pattern Recognition,Part XII》 - International Conference on Pattern Recognition - 2025, - 235～250 - 共16页

摘要： of epileptic biomarker detection, potentially | Accurate estimation of the epileptogenic zone | is critical for surgical treatment of drug | -resistant epilepsy. While epilepsy biomarkers detection | traditionally relied on expert clinician knowledge, the use of

关键词： Multi-modal | Multitask learning (MTL) | High frequency oscillation (HFO) | Interictal epileptiform discharges (IED) | Epilepsy

8. CARB-Net: Camera-Assisted Radar-Based Network for Vulnerable Road User Detection NSTL国家科技图书文献中心

Wei-Yu Lee | Martin Dimitrievski ... - 《Computer Vision - ECCV 2024,Part LXIV》 - European Conference on Computer Vision - 2025, - 294～310 - 共17页

摘要： stages of a radar-based detection* network. Furthermore* | Ensuring a reliable perception of vulnerable | . Radar stands out* as an appealing sensor choice due to* | merges the angular accuracy of a camera with the | robustness and depth sensing capabilities of radar. We

关键词： Micro-Doppler signature | Vulnerable road user detection

9. Predicting Surgical Site Infections: A Time to Event Approach NSTL国家科技图书文献中心

Tiago Lopes | Julio Duarte ... - 《Progress in Artificial Intelligence,Part II》 - EPIA Conference on Artificial Intelligence - 2025, - 50～61 - 共12页

摘要：-consuming, resulting in late detection* of this pathology* | detection of infections resulting from surgical | Nosocomial infections are problems of enormous | proportions that impact the well-being of hospitalized | in ensuring better quality of life for patients and

关键词： Surgical site infections | Machine learning | Artificial intelligence | ABI

10. Test beam results on 3D pixel sensors for the CMS Tracker upgrade at the High-Luminosity LHC NSTL国家科技图书文献中心

Lasaosa, C. - 《Journal of instrumentation》 - 2025,20(3) - 共10页

摘要： been carried out* in the context of the CMS tracker R* | The High Luminosity upgrade of the CERN Large | tolerant silicon pixel sensors for the innermost part of | innermost layer of the tracker, which is as close as 3 cm | fluence of 3.4 x 1016 neqcm-2 during the high-luminosity

摘要： networks to fraud detection* still faces challenges of* | capturing two types of fraudulent behaviors: the context | encoding module which encodes the time intervals of out | address the challenges, we present Context-encoding-and | GNN-based fraud detection* model. To capture the*

关键词： Camouflage | Fraud detection | Graph neural network

摘要： constructed to filter high-quality out* of distribution* | The distribution of actual industrial process | data follows a Gaussian distribution. In this context | , we propose a novel fault detection* method based on* | improved supervised contrastive learning detection* model*

关键词： Fault detection | Denoising diffusion probabilistic model | Supervised contrastive learning | Out-of-distribution data | Process safety

摘要： enhancing detection* performance. Using a dataset of 3000* | effectiveness of CNN-based malware detection. This work stands | out by focusing on the dynamic tuning of deep | malware detection, current approaches face notable | , achieving high detection* accuracy, and maintaining*

关键词： Android | Malware detection | Hyperparameters tuning | Convolutional neural network | Pre-trained model

摘要： the left atrial appendage. The detection* of thrombi* | complications-especially in the context* of rhythm control. The* | morphology. This structure is a preferred region of | of the atrial wall. Blood stasis occurs primarily | in low flow conditions. One of the tasks of

关键词： Atrial fibrillation | Echocardiography | Left atrial appendage | LAA morphology | LAA function | LAA intervention

摘要：Studies on behavioural ecology of tadpoles of | morphological features. 2) The sensory basis of food or | predator detection* in herbivorous tadpoles is* | visual and chemical cues in prey detection/ capture. 3 | (kairomones, alarm cues or dietary metabolites of

关键词： Antipredator defences | Developmental plasticity | Kin recognition | Microhabitat choice | Predator-prey interactions | Tadpoles

摘要：Multimodal remote sensing object detection* (MM* | design of channel-gated spatial cross-attention (CSCA | -modal context* interaction through static convolution* | kernels to filter out* irrelevant information between* | -specific global context* to fuse features, achieving*

关键词： Feature extraction | Remote sensing | Object detection | Termination of employment | Redundancy | Optical imaging | Optical filters | Filters | Fuses | Uncertainty

摘要： of epileptic biomarker detection, potentially | Accurate estimation of the epileptogenic zone | is critical for surgical treatment of drug | -resistant epilepsy. While epilepsy biomarkers detection | traditionally relied on expert clinician knowledge, the use of

关键词： Multi-modal | Multitask learning (MTL) | High frequency oscillation (HFO) | Interictal epileptiform discharges (IED) | Epilepsy

摘要： stages of a radar-based detection* network. Furthermore* | Ensuring a reliable perception of vulnerable | . Radar stands out* as an appealing sensor choice due to* | merges the angular accuracy of a camera with the | robustness and depth sensing capabilities of radar. We

关键词： Micro-Doppler signature | Vulnerable road user detection

摘要：-consuming, resulting in late detection* of this pathology* | detection of infections resulting from surgical | Nosocomial infections are problems of enormous | proportions that impact the well-being of hospitalized | in ensuring better quality of life for patients and

关键词： Surgical site infections | Machine learning | Artificial intelligence | ABI

摘要： been carried out* in the context of the CMS tracker R* | The High Luminosity upgrade of the CERN Large | tolerant silicon pixel sensors for the innermost part of | innermost layer of the tracker, which is as close as 3 cm | fluence of 3.4 x 1016 neqcm-2 during the high-luminosity

关键词： Hybrid detectors | Particle tracking detectors (Solid-state detectors) | Radiation-hard detectors

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： networks to fraud detection still faces challenges of | capturing two types of fraudulent behaviors: the context | encoding module which encodes the time intervals of out | address the challenges, we present Context-encoding-and | GNN-based fraud detection model. To capture the

关键词： Camouflage | Fraud detection | Graph neural network

摘要： constructed to filter high-quality out of distribution | The distribution of actual industrial process | data follows a Gaussian distribution. In this context | , we propose a novel fault detection method based on | improved supervised contrastive learning detection model

关键词： Fault detection | Denoising diffusion probabilistic model | Supervised contrastive learning | Out-of-distribution data | Process safety

摘要： enhancing detection performance. Using a dataset of 3000 | effectiveness of CNN-based malware detection. This work stands | out by focusing on the dynamic tuning of deep | malware detection, current approaches face notable | , achieving high detection accuracy, and maintaining

关键词： Android | Malware detection | Hyperparameters tuning | Convolutional neural network | Pre-trained model

摘要： the left atrial appendage. The detection of thrombi | complications-especially in the context of rhythm control. The | morphology. This structure is a preferred region of | of the atrial wall. Blood stasis occurs primarily | in low flow conditions. One of the tasks of

关键词： Atrial fibrillation | Echocardiography | Left atrial appendage | LAA morphology | LAA function | LAA intervention

摘要：Studies on behavioural ecology of tadpoles of | morphological features. 2) The sensory basis of food or | predator detection in herbivorous tadpoles is | visual and chemical cues in prey detection/ capture. 3 | (kairomones, alarm cues or dietary metabolites of

关键词： Antipredator defences | Developmental plasticity | Kin recognition | Microhabitat choice | Predator-prey interactions | Tadpoles

摘要：Multimodal remote sensing object detection (MM | design of channel-gated spatial cross-attention (CSCA | -modal context interaction through static convolution | kernels to filter out irrelevant information between | -specific global context to fuse features, achieving

关键词： Feature extraction | Remote sensing | Object detection | Termination of employment | Redundancy | Optical imaging | Optical filters | Filters | Fuses | Uncertainty

摘要： of epileptic biomarker detection, potentially | Accurate estimation of the epileptogenic zone | is critical for surgical treatment of drug | -resistant epilepsy. While epilepsy biomarkers detection | traditionally relied on expert clinician knowledge, the use of

关键词： Multi-modal | Multitask learning (MTL) | High frequency oscillation (HFO) | Interictal epileptiform discharges (IED) | Epilepsy

摘要： stages of a radar-based detection network. Furthermore | Ensuring a reliable perception of vulnerable | . Radar stands out as an appealing sensor choice due to | merges the angular accuracy of a camera with the | robustness and depth sensing capabilities of radar. We

关键词： Micro-Doppler signature | Vulnerable road user detection

摘要：-consuming, resulting in late detection of this pathology | detection of infections resulting from surgical | Nosocomial infections are problems of enormous | proportions that impact the well-being of hospitalized | in ensuring better quality of life for patients and

关键词： Surgical site infections | Machine learning | Artificial intelligence | ABI

摘要： been carried out in the context of the CMS tracker R | The High Luminosity upgrade of the CERN Large | tolerant silicon pixel sensors for the innermost part of | innermost layer of the tracker, which is as close as 3 cm | fluence of 3.4 x 1016 neqcm-2 during the high-luminosity

关键词： Hybrid detectors | Particle tracking detectors (Solid-state detectors) | Radiation-hard detectors

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： networks to fraud detection* still faces challenges of* | capturing two types of fraudulent behaviors: the context | encoding module which encodes the time intervals of out | address the challenges, we present Context-encoding-and | GNN-based fraud detection* model. To capture the*

摘要： constructed to filter high-quality out* of distribution* | The distribution of actual industrial process | data follows a Gaussian distribution. In this context | , we propose a novel fault detection* method based on* | improved supervised contrastive learning detection* model*

摘要： enhancing detection* performance. Using a dataset of 3000* | effectiveness of CNN-based malware detection. This work stands | out by focusing on the dynamic tuning of deep | malware detection, current approaches face notable | , achieving high detection* accuracy, and maintaining*

摘要： the left atrial appendage. The detection* of thrombi* | complications-especially in the context* of rhythm control. The* | morphology. This structure is a preferred region of | of the atrial wall. Blood stasis occurs primarily | in low flow conditions. One of the tasks of

摘要：Studies on behavioural ecology of tadpoles of | morphological features. 2) The sensory basis of food or | predator detection* in herbivorous tadpoles is* | visual and chemical cues in prey detection/ capture. 3 | (kairomones, alarm cues or dietary metabolites of

摘要：Multimodal remote sensing object detection* (MM* | design of channel-gated spatial cross-attention (CSCA | -modal context* interaction through static convolution* | kernels to filter out* irrelevant information between* | -specific global context* to fuse features, achieving*

摘要： stages of a radar-based detection* network. Furthermore* | Ensuring a reliable perception of vulnerable | . Radar stands out* as an appealing sensor choice due to* | merges the angular accuracy of a camera with the | robustness and depth sensing capabilities of radar. We

摘要：-consuming, resulting in late detection* of this pathology* | detection of infections resulting from surgical | Nosocomial infections are problems of enormous | proportions that impact the well-being of hospitalized | in ensuring better quality of life for patients and

摘要： been carried out* in the context of the CMS tracker R* | The High Luminosity upgrade of the CERN Large | tolerant silicon pixel sensors for the innermost part of | innermost layer of the tracker, which is as close as 3 cm | fluence of 3.4 x 1016 neqcm-2 during the high-luminosity