A5101634508- Neural Networks and Applications
- Topic Modeling
- Bioinformatics and Genomic Networks
- Metabolomics and Mass Spectrometry Studies
- Artificial Intelligence in Healthcare
- Advanced Graph Neural Networks
- Cell Image Analysis Techniques
- Natural Language Processing Techniques
- Advanced MRI Techniques and Applications
- Multimodal Machine Learning Applications
- Machine Learning and Data Classification
- Domain Adaptation and Few-Shot Learning
- Metallurgy and Material Forming
- Image Retrieval and Classification Techniques
- Alzheimer's disease research and treatments
- Thyroid Disorders and Treatments
- Primate Behavior and Ecology
- Advanced Text Analysis Techniques
- Recommender Systems and Techniques
- Remote Sensing and LiDAR Applications
- Internet Traffic Analysis and Secure E-voting
- Human Pose and Action Recognition
- Terahertz technology and applications
- Advanced Optical Sensing Technologies
- Machine Learning in Bioinformatics
University of Electronic Science and Technology of China
2021-2025
Beijing University of Posts and Telecommunications
2022-2024
Harbin Engineering University
2021-2024
Beijing Anzhen Hospital
2023
Nanjing University of Aeronautics and Astronautics
2022
Huazhong Agricultural University
2022
Central China Normal University
2021
The development of autonomous vehicles and unmanned aerial has led to a current research focus on improving the environmental perception automation equipment. platform detects its surroundings then makes decision based information. major challenge is detect classify objects precisely; thus, it necessary perform fusion different heterogeneous data achieve complementary advantages. In this paper, robust object detection classification algorithm millimeter-wave (MMW) radar camera proposed....
Federated learning (FL) has gained widespread attention for its privacy-preserving and collaborative capabilities. Due to significant statistical heterogeneity, traditional FL struggles generalize a shared model across diverse data domains. Personalized federated addresses this issue by dividing the into globally part locally private part, with local correcting representation biases introduced global model. Nevertheless, converged parameters more accurately capture domain-specific knowledge,...
Testosterone is essential to human growth and development as well immune regulation. Zika virus (ZIKV), an emerging arbovirus associated with neurological complications including neuroinflammation, can also cause testicular damage decrease testosterone secretion. However, whether the dysregulation of plays a role in process neuroinflammation during ZIKV pathogenesis still unclear. In this study, we found that infection caused decreased secretion male mice, supplementation after reduced their...
The distinguishable subregions that compose the hippocampus are differently involved in functions associated with Alzheimer’s disease (AD). Thus, identification of hippocampal and genes classify AD healthy control (HC) groups high accuracy is meaningful. In this study, by jointly analyzing multimodal data, we propose a novel method to construct fusion features classification based on random forest for identifying important features. Specifically, using gene sequence correlation reduce...
Abstract High detail and fast magnetic resonance imaging (MRI) sequences are highly demanded in clinical settings, as inadequate information can lead to diagnostic difficulties. MR image super-resolution (SR) is a promising way address this issue, but its performance limited due the practical difficulty of acquiring paired low- high-resolution (LR HR) images. Most existing methods generate these pairs by down-sampling HR images, process that often fails capture complex degradations...
Precipitation regime has been dramatically shifted under global warming, and will especially reshape the water-limited dryland ecosystems, biomes therein. Among biological organisms, biocrust (complex communities consisting of soil particles, photoautotrophic (algae, cyanobacteria, lichens, mosses), heterotrophic (e.g., bacteria, fungi, archaea) organisms) is a critical component, tends to be fragile relative vascular plants. However, it remains largely unclear how do different types respond...
Landmark-based 3-hop cover labeling is a category of approaches for shortest distance/path queries on large-scale complex networks. It pre-computes an index offline to accelerate the online query. Most real-world graphs undergo rapid changes in topology, which makes maintenance dynamic necessary. So far, majority methods can handle only one edge update (either addition or deletion) each time. To keep up with frequently changing graphs, we research ful ly b atch m aintenance problem labeling,...
Domain-Specific Chinese Relation Extraction (DSCRE) aims to extract relations between entities from domain-specific text. Despite the rapid development of PLMs in recent years, especially LLMs, DSCRE still faces three core challenges: complex network structure design, poor awareness, and high consumption fine-tuning. Given impressive performance large language models (LLMs) natural processing, we propose a new framework called CRE-LLM. This is based on fine-tuning open-source such as...
Brain transcriptomics provides insights into the molecular mechanisms by which brain coordinates its functions and processes. However, existing multimodal methods for predicting Alzheimer's disease (AD) primarily rely on imaging sometimes genetic data, often neglecting transcriptomic basis of brain. Furthermore, while striving to integrate complementary information between modalities, most studies overlook informativeness disparities modalities. Here, we propose TMM, a trusted multiview...
The key challenge of cross-modal domain-incremental learning (DIL) is to enable the model continuously learn from novel data with different feature distributions under same task without forgetting old ones. However, existing top-performing methods still cause high rates, by lacking intra-domain knowledge extraction and inter-domain common prompting strategy. In this paper, we propose a simple yet effective framework, CP-Prompt, training limited parameters instruct pre-trained new domains...
The distinct characteristics of multiomics data, including complex interactions within and across biological layers disease heterogeneity (e.g., in etiology clinical symptoms), drive us to develop novel designs address unique challenges prediction. In this paper, we propose the multi-view knowledge transfer learning (MVKTrans) framework, which transfers intra- inter-omics an adaptive manner by reviewing data suppressing bias transfer, thereby enhancing classification performance....
Abstract Background: IDH is a common and serious complication of chronic hemodialysis that can lead to adverse long-term outcomes, including increased cardiovascular all-cause mortality. Predicting in advance may aid establishing an intervention plan. Method: This study used 1173783 sessions from 837 patients 211 features (including time-span varying basic information features) as datasets. We constructed 21 ion changes before after dialysis) by clinical physicians the raw selected 89 grid...
Temporal knowledge graph (TKG) has been proved to be an effective way for modeling dynamic facts in real world. Many efforts have devoted into predicting future events i.e. extrapolation, on TKGs. Recently, rule-based completion methods which are considered more interpretable than embedding-based methods, transferred temporal extrapolation. However, models suffer from redundancy when leveraged under settings, results inaccurate rule confidence calculation. In this paper, we define the...