A5100431708- Advanced Computational Techniques and Applications
- Service-Oriented Architecture and Web Services
- Advanced Neural Network Applications
- Anomaly Detection Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Domain Adaptation and Few-Shot Learning
- Advanced Clustering Algorithms Research
- COVID-19 diagnosis using AI
- Advanced Text Analysis Techniques
- Complex Network Analysis Techniques
- Image Retrieval and Classification Techniques
- Algorithms and Data Compression
- Genomics and Phylogenetic Studies
- Advanced MRI Techniques and Applications
- Advanced Neuroimaging Techniques and Applications
- Network Security and Intrusion Detection
- Semantic Web and Ontologies
- COVID-19 Clinical Research Studies
- Medical Imaging and Analysis
- Innovation and Knowledge Management
- Machine Learning and ELM
- Advanced Malware Detection Techniques
- Hydraulic flow and structures
- Software Engineering Research
- Hydraulic Fracturing and Reservoir Analysis
Beijing Information Science & Technology University
2025
Mayo Clinic in Florida
2025
WinnMed
2025
Changchun University of Science and Technology
2020-2025
Tongji University
2013-2024
Ningbo Institute of Industrial Technology
2024
Chinese Academy of Sciences
2024
Shanghai Tenth People's Hospital
2016-2024
Shandong Provincial Hospital
2024
Shandong First Medical University
2024
Multi-view deep learning is performed based on the fusion of data from multiple sources, i.e. with views. However, due to property differences and inconsistency results multi-view may be uncertain unreliable. It required reduce uncertainty in implement trusted learning. Aiming at problem, we revisit perspective opinion aggregation thereby devise a method. Within this method, adopt evidence theory formulate opinions as different sources measure through accumulation. We prove that accumulating...
Accurate virtual orbital reconstruction is crucial for preoperative planning. Traditional methods, such as the mirroring technique, are unsuitable defects involving both sides of midline and time-consuming labor-intensive. This study introduces a modified 3D U-Net+++ architecture reconstruction, aiming to enhance precision automation. The model was trained tested with 300 synthetic from cranial spiral CT scans. method validated in 15 clinical cases fractures evaluated using quantitative...
Multi-view deep classification expects to obtain better performance than using a single view. However, due the uncertainty and inconsistency of data sources, adding views does not necessarily lead improvements in multi-view classification. How avoid worsening when is crucial for learning but rarely studied. To tackle this limitation, paper, we reformulate problem from perspective safe thereby propose Safe Deep Classification (SMDC) method, which can guarantee that deteriorate fusing multiple...
The cell wall is a dynamic structure that important for the pathogenicity of Candida albicans. Mannan, which located in outermost layer wall, has been shown to contribute pathogenesis C. albicans, however, molecular mechanism by this occurs remains unclear. Here we identified novel α-1,6-mannosyltransferase encoded MNN10 We found Mnn10 required α-1,6-mannose backbone biosynthesis and polysaccharides organization. Deletion resulted significant attenuation albicans murine systemic candidiasis...
Purpose This paper aims to use a concurrent mixed method approach explore the key variables that can influence customer experience at food and wine event. Design/methodology/approach A methods approach, using participant-generated image (PGI) method, together with recall survey, provided images associated narratives, descriptive statistics, correlations hierarchical multiple regression analysis how attendees appraise their experiences based on goals link between appraisals overall...
Most of the traditional classification algorithms are based on premise that datasets uniformly distributed or roughly equivalent. Once sample dataset is not balanced , performance drops sharply. To efficiently deal with imbalance data, an improved generative adversarial network (GAN) algorithm proposed in this work . Firstly, we construct artificial samples so more minority-class's data can be obtained via optimizing GAN loss function. Secondly, build a fully-connected for structured...
Fine-Grained Image Classification (FGIC) aims to classify images into specific subordinate classes of a superclass. Due insufficient training data and confusing samples, FGIC may produce uncertain classification results that are untrusted for applications. In fact, can be viewed as hierarchical process the multilayer information facilitates reduce uncertainty improve reliability FGIC. this paper, we adopt evidence theory measure confidence in propose trusted method through fusing evidence....
The formation of the secondary phases in 2205 duplex stainless steel (DSS) influences greatly its mechanical properties. Studies on microstructural evolution and coarsening behavior precipitates DSS have both scientific technological significance. In present work, composition morphology aged at 850◦C for periods up to 200h are systematically investigated by SEM/EDS (Scanning Electron Microscope/ Energy Dispersive Spectrometer) TEM (Transmission Microscope). addition, size statistics...
"Rebooting Computing" (RC) is an effort in the IEEE to rethink future computers. RC started 2012 by co-chairs, Elie Track (IEEE Council on Superconductivity) and Tom Conte (Computer Society). takes a holistic approach, considering revolutionary as well evolutionary solutions needed advance computer technologies. Three summits have been held 2013 2014, discussing different technologies, from emerging devices user interface, security energy efficiency, neuromorphic reversible computing. The...
Keyphrase Prediction (KP) task aims at predicting several keyphrases that can summarize the main idea of given document.Mainstream KP methods be categorized into purely generative approaches and integrated models with extraction generation.However, these either ignore diversity among or only weakly capture relation across tasks implicitly.In this paper, we propose UniKeyphrase, a novel end-to-end learning framework jointly learns to extract generate keyphrases.In stacked layer bagof-words...
Many human-centric systems have begun to use business process management technology in production. With the operation of systems, more and logs data been accumulated. However, effective utilization analysis these event are challenges that people need solve urgently. Process mining is a branch technology. It can extract knowledge from build models, which helps detect improve processes. The current algorithms inadequate dealing with log noise. family alpha-algorithms ignores impact noise,...
Aircraft detection in remote sensing images plays an essential role military and civil applications. Many aircraft are small targets covered by cloud fog, which make it difficult to extract sufficient feature information for detection. In this paper, a single-stage object model YOLM (You can Look More) has achieved better performance extracting more features. includes enhanced neck network with four-layer pyramid path aggregation network. More layers enable the detailed information. To use...
The development of object detection plays an important role in the realisation fully autonomous driving, and feature extraction is key step for detection. There has been significant difference scale variation features different road traffic participants (RTPs), meanwhile traditional Convolutional Neural Networks (CNNs) was difficult to extract efficiently small targets. In order improve ability extraction, a RTP method combining dynamic convolution enhancement proposed. Fully One-Stage...
A new method has been developed to reduce the number of phase-encoding steps in a multi-echo spin-echo imaging sequence allowing fast T(2) mapping without loss spatial resolution. In proposed approach, k-space data at each echo time were undersampled and reconstruction algorithm that exploited temporal correlation MR signal was used reconstruct alias-free images. specific application this with multiple-receiver acquisition, offering an alternative existing parallel methods, also introduced....
In a large-scale wireless sensor network (WSN), densely distributed nodes process large amount of data. The aggregation data in can consume great energy. To balance and reduce the energy consumption WSN extend life, this paper proposes nonuniform clustering routing algorithm based on improved K-means algorithm. uses method to form optimize clusters, it selects appropriate cluster heads life cycle WSN. ensure that head (CH) selection is fair location selected CH not concentrated within...
Our previous studies indicated that Yiguanjian decoction (YGJ) has an anti-hepatic-fibrosis effect and could regulate macrophage status.To elucidate the mechanism of YGJ in regulating macrophages.Liver cirrhosis was induced by CCl4 for 12 weeks combined with 2-acetylaminofluorene (2-AAF) last 4 male Wistar rats. (3.56 mg/kg) orally administered weeks, SORA (1 as control. In vitro, RAW264.7 cells were treated lipopolysaccharides (LPSs) to induce polarization M1 phenotype, they co-cultured...