A5073112133- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Visual Attention and Saliency Detection
- HIV Research and Treatment
- Advanced Vision and Imaging
- Infrastructure Maintenance and Monitoring
- Industrial Vision Systems and Defect Detection
- CAR-T cell therapy research
- Image and Video Quality Assessment
- Acute Myeloid Leukemia Research
- Ubiquitin and proteasome pathways
- Cytomegalovirus and herpesvirus research
- Immunotherapy and Immune Responses
- Immune Cell Function and Interaction
- Image Processing Techniques and Applications
- Railway Engineering and Dynamics
- Immunodeficiency and Autoimmune Disorders
- Human Mobility and Location-Based Analysis
- Currency Recognition and Detection
- Immune Response and Inflammation
- Ocular and Laser Science Research
- Autonomous Vehicle Technology and Safety
- Advanced Biosensing Techniques and Applications
- Electrical Contact Performance and Analysis
Children's Hospital of Zhejiang University
2018-2024
Zhejiang University
2010-2024
Zhejiang University of Science and Technology
2024
First Affiliated Hospital Zhejiang University
2011-2012
State Key Laboratory of Diagnosis and Treatment of Infectious Diseases
2010-2012
Rail surface defect detection for traffic safety has received considerable attention. With the development of deep learning, numerous methods combining RGB and depth information have been proposed. However, these directly fuse raw features extracted from backbone, which can lead to ineffective use complementary two modalities. In this study, we developed a contour semantic feature alignment fusion network (CSANet) with bidirectional explore internal consistency cross-modal both perspectives....
Crowd density estimation has gained significant research interest owing to its potential in various industries and social applications. Therefore, this paper proposes a multistyle joint-perception network based on knowledge distillation-trained student (MJPNet-S*) for drone-based red–green–blue, thermal/depth (RGB-T/D) crowd tasks. To provide superior accuracy efficiency, novel trimodal working module effectively combines the modalities facilitate comprehensive extraction utilization. A...
Significant advancements in RGB-D semantic segmentation have been made owing to the increasing availability of robust depth information. Most researchers combined with RGB data capture complementary information images. Although this approach improves performance, it requires excessive model parameters. To address problem, we propose DGPINet-KD, a deep-guided and progressive integration network knowledge distillation (KD) for indoor scene analysis. First, used branching attention guidance...
Color–thermal (RGB-T) urban scene parsing has recently attracted widespread interest. However, most existing approaches to RGB-T do not deeply explore the information complementarity between features. In this study, we propose a cross-modal attention-cascaded fusion network (CACFNet) that fully exploits cross-modality. our design, attention module mines complementary from two modalities. Subsequently, cascaded decodes multi-level features in an up-bottom manner. Noting each pixel is labeled...
Crowd counting has received significant attention in recent years due to its practical applications. In order address the specific characteristics of RGB and thermal images, we have developed graph enhancement transformer aggregation network (GETANet) for generating representative density maps. Our approach incorporates several innovative modules enhance accuracy. Firstly, introduced a position-adaptive module that effectively counts individuals' positions integrates features extracted from...
Owing to extensive research on deep learning, significant progress has recently been made in trackless surface defect detection (SDD). Nevertheless, existing algorithms face two main challenges. First, while depth features contain rich spatial structure features, most models only accept red-green-blue (RGB) as input, which severely constrains performance. Thus, this study proposes a dual-stream teacher model termed the asymmetrical contrastive learning network (ACLNet-T), extracts both RGB...
To determine the efficacy and safety of eltrombopag (E-PAG) combined with intensive immunosuppressive therapy (IST) for treatment pediatric patients severe aplastic anemia (SAA). A total 57 newly diagnosed were enrolled in this study. Thirty nine treated IST alone, consisting porcine anti-human thymocyte globulin (30 mg/kg/day × 5 days) cyclosporine (CsA) (treated 2 years, a trough concentration maintained at 200-250 ng/mL), 18 + E-PAG (12.5-50 mg/day, 6 months). We found no statistical...
In recent years, fused infrared and visible-light computer vision methods based on fully convolutional networks (FCNs) have achieved remarkable results due to their complementary features. However, the growing requirements for tasks continually demand an increasing number of parameters, leading computational costs that are considerably high deployment mobile devices. Hence, building a small capable lightweight student model supervised trained by large teacher with superior performance...
Dendritic cell-specific intercellular adhesion molecule-3 grabbing nonintegrin (DC-SIGN) is an important pattern recognition receptor on dendritic cells (DCs), and its expression shows significant cytological histological specificity, being interleukine-4 (IL-4) dependent. The signaling pathways through which IL-4 regulates of DC-SIGN are still unclear. We used phorbol 12-myristate 13-acetate- (PMA-) differentiated THP-1 as the in vitro model monocyte/macrophage to study involved...
In this study, we investigated the mechanisms underlying anti-inflammatory effects of honokiol in tumor necrosis factor (TNF)-α-stimulated rheumatoid arthritis synovial fibroblasts (RASFs). RASFs pre-treated with (0-20 μM) were stimulated TNF-α (20 ng/ml). The levels prostaglandin E2 (PGE2), nitric oxide (NO), soluble intercellular adhesion molecule-1 (sICAM-1), transforming growth factor-β1 (TGF-β1), monocyte chemotactic protein-1 (MCP-1), and macrophage inflammatory protein-1α (MIP-1α)...
Significant progress has been made in the field of indoor scene parsing. The increasing demand for lightweight networks is due to limited hardware capacity mobile devices. However, there a lack research on design Therefore, we propose dual stream network (LDSNet) with knowledge distillation (KD) RGB-D Initially, developed two-stream three versions (LDSNet-tiny*, LDSNet-small*, and LDSNet-base, where * represents model after KD) different scenarios. In main stream, designed an integrated...
In recent years, optical remote sensing image (ORSI) scene analysis has attracted increasing interest. However, existing networks show a trend of bifurcation. Lightweight have very high inference speed but poor contextual information in highly complex backgrounds. contrast, with high-performance reasoning capability require many parameters and are computationally expensive. Since the knowledge distillation method can greatly lighten model, we propose graph semantic guided network (GSGNet)...
Here, we introduced the first case of acute myeloid leukemia (AML) with RARG-NUP98 in a pediatric patient. The young male presented structural and functional abnormalities similar to hypergranular promyelocytic leukemia, but was resistant all transretinoic acids arsenic trioxide. Till date, only 12 adult AML cases involving RARG rearrangement have been reported. At present, there is no standardized or optimal treatment option for this subtype. Disease management may typically require joint...
Transmission line (TL) inspection plays a crucial role in maintaining reliable electricity supply to all regions. Computer vision methods, especially those utilizing infrared images, have achieved significant advancements this field. However, many existing multimodal fusion methods utilize conventional attention mechanisms or simple meta-additions combine different modalities without proper alignment. Moreover, these often rely on large number of parameters achieve better performance. To...
CD19 chimeric antigen receptor T (CD19CAR-T) cell therapy has shown striking response in treating relapsed and refractory B-lineage acute lymphoblastic leukemia (r/r B-ALL). However, side-effects including cytokine release syndrome (CRS) neurotoxicity can be fatal to patients. In this report, five patients with r/r B-ALL were treated CD19CAR-T cells. Cytokine experienced by four who achieved complete remission (CR) minimal residual disease (MRD) negative. One patient did not the treatment...
Plasmacytoid dendritic cells (pDCs), not only inhibit viral replication, but also play an essential role in linking the innate and adaptive immune system. In this study, we explored effects of human immunodeficiency virus (HIV) gp120 tat on CpG-A-induced inflammatory cytokines pDCs. The results provided fundamental insights into HIV pathogenesis that may hold promise for preventative even curative strategies. pDCs were isolated using blood DC antigen 4 (BDCA-4) isolation kit, purity was...