A5100635998- Advanced Memory and Neural Computing
- Image and Video Quality Assessment
- Neuroscience and Neural Engineering
- Advanced Image Fusion Techniques
- Traffic and Road Safety
- Visual Attention and Saliency Detection
- Hate Speech and Cyberbullying Detection
- Traffic Prediction and Management Techniques
- Blind Source Separation Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- Intraocular Surgery and Lenses
- Ferroelectric and Negative Capacitance Devices
- Advanced Adaptive Filtering Techniques
- Retinal and Macular Surgery
- Statistical Methods and Inference
- Multimodal Machine Learning Applications
- Color Science and Applications
- CCD and CMOS Imaging Sensors
- Control Systems and Identification
- Image Enhancement Techniques
- Semiconductor materials and devices
- Vascular Malformations Diagnosis and Treatment
- Speech and Audio Processing
- Maritime Navigation and Safety
- Autonomous Vehicle Technology and Safety
Chinese Academy of Medical Sciences & Peking Union Medical College
2025
Concordia University
2024
Beijing Tongren Hospital
2023-2024
Capital Medical University
2023-2024
Eye Center
2024
State Key Laboratory of Respiratory Disease
2024
Guangzhou Medical University
2024
Gannan Medical University
2024
Oxford Brookes University
2016-2024
Xinjiang Normal University
2024
Blind image quality assessment (BIQA) methods aim to predict of images as perceived by humans without access a reference image. Recently, deep learning have gained substantial attention in the research community and proven useful for BIQA. Although previous study neural networks (DNN) is presented, some novelty DNN methods, which are recently proposed, not summarized In this paper, we provide survey covering various First, systematically analyze existing DNN-based according role DNN. Then,...
AXL, a receptor tyrosine kinase, has recently emerged as potential therapeutic target against various types of cancer. Gilteritinib, FDA-approved small-molecule inhibitor, is used for the treatment patients with FLT3-mutated acute myeloid leukemia. However, antitumor activity gilteritinib in solid tumors remains poorly elucidated. In this study, we explored AXL-expressing esophageal cancer (EC), ovarian (OC), and gastric (GC), along underlying molecular mechanisms. Our data demonstrated that...
Image quality assessment (IQA) based on deep learning faces the overfitting problem due to limited training samples available in existing IQA databases. Transfer is a plausible solution problem, which shared features derived from large-scale Imagenet source domain could be transferred original recognition task intended task. However, and target as well their corresponding tasks are not directly related. In this paper, we propose new transitive transfer method for no-reference image...
Blind/no-reference image quality assessment is performed to accurately evaluate the perceptual of a distorted without prior information from reference image. In this paper, an effective blind approach based on entropy differences in discrete cosine transform domain for natural images proposed. Information measure amount We find coefficient distribution shows pulse-shape phenomenon, which directly affects entropy. Then, Weibull model used fit distributions and images. This because...
Crassulacean acid metabolism (CAM) has high water-use efficiency (WUE) and is widely recognized to have evolved from C3 photosynthesis. Different plant lineages convergently CAM, but the molecular mechanism that underlies C3-to-CAM evolution remains be clarified. Platycerium bifurcatum (elkhorn fern) provides an opportunity study changes underlying transition CAM photosynthesis because both modes of occur in this species, with sporotrophophyll leaves (SLs) cover (CLs) performing weak...
Exploring and analyzing safety influencing factors can guide targeted traffic management. Traditional models are aimed at specific data problems making adjustments to the model structure, which lack focus on predictive ability have limited information analysis of factors. In recent years, machine-learning methods opened new avenues in modeling that higher prediction accuracy, identify complex nonlinear relationships, overcome over- under-dispersion correlation. Machine-learning methods,...
Coreference resolution – which is a crucial task for understanding discourse and language at large has yet to witness widespread benefits from models (LLMs). Moreover, coreference systems largely rely on supervised labels, are highly expensive difficult annotate, thus making it ripe prompt engineering. In this paper, we introduce QA-based prompt-engineering method discern generative, pre-trained LLMs' abilities limitations toward the of resolution. Our experiments show that GPT-2 GPT-Neo can...
Online sexism has become an increasing concern in social media platforms as it affected the healthy development of Internet and can have negative effects society. While research detection domain is growing, most this focuses on English language Twitter platform. Our objective here to broaden scope by considering Chinese Sina Weibo. We propose first dataset -- Weibo Sexism Review (SWSR) --, well a large lexicon SexHateLex made abusive gender-related terms. introduce our data collection...
The quantised kernel least mean square (QKLMS) is a simple yet efficient online learning algorithm, which reduces the computational cost significantly by quantising input space to constrain growth of network size. QKLMS considers only compression and assumes that desired outputs data are equal those closest centres. In many cases, however, in neighbourhood may have big differences, especially when underlying system disturbed impulsive noises. Such fluctuation seriously deteriorate...
Based on big data and artificial intelligence technology, the judicial intelligent service platform provides users with efficient accurate legal consulting services. It has many functions such as question answer, case analysis, law inquiry. also analyzes application value of in field data. By building an platform, rapid processing analysis massive been realized, efficiency quality work have significantly improved. In addition, it emphasizes importance management control data, puts forward...
The prevailing design philosophy for polymeric vectors delivering siRNA is rooted in the post-transcriptional gene silencing (PTGS) mechanism. Yet, transcriptional (TGS) mechanism offers a potentially more durable effect, which necessitates efficient delivery into nucleus. However, it remains challenge to efficiently deliver We have explored guanidinylated cyclic synthetic polypeptides (GCSPs) enhance nuclear of siRNA, but an increased cytotoxicity and difficulty producing GCSPs on large...
Error or drift is frequently produced in pose estimation based on geometric "feature detection and tracking" monocular visual odometry(VO) when the speed of camera movement exceeds 1.5m/s. While, most VO methods deep learning, weight factors are form fixed values, which easy to lead overfitting. A new measurement system, for odometry, named Deep Learning Visual Odometry(DLVO), proposed neural network. In this Convolutional Neural Network(CNN) used extract feature perform matching. Moreover,...
One challenge facing image quality assessment (IQA) is that current models designed or trained on the basis of exiting databases are intrinsically suboptimal and cannot deal with real-world complexity diversity natural scenes. IQA heavily skewed toward visibility distortions. It critical to understand wider determinants perceived use new understanding improve predictive power models. Human behavioral categorization performance powerful essential for visual tasks. However, little known about...
Research on image quality assessment (IQA) remains limited mainly due to our incomplete knowledge about human visual perception. Existing IQA algorithms have been designed or trained with insufficient subjective data a small degree of stimulus variability. This has led challenges for those handle complexity and diversity real-world digital content. Perceptual evidence from subjects serves as grounding the development advanced algorithms. It is thus critical acquire reliable controlled...