A5100449952- Cryptography and Data Security
- Radiomics and Machine Learning in Medical Imaging
- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- AI in cancer detection
- Medical Imaging and Analysis
- Topic Modeling
- Advanced X-ray and CT Imaging
- Advanced DC-DC Converters
- Medical Imaging Techniques and Applications
- Advanced Vision and Imaging
- Silicon Carbide Semiconductor Technologies
- Image Enhancement Techniques
- Natural Language Processing Techniques
- Cryptographic Implementations and Security
- Cryptography and Residue Arithmetic
- Advanced Image Processing Techniques
- Cloud Data Security Solutions
- Brain Tumor Detection and Classification
- Image and Signal Denoising Methods
- Advanced MRI Techniques and Applications
- Handwritten Text Recognition Techniques
- Multimodal Machine Learning Applications
- Electromagnetic Compatibility and Noise Suppression
- Radiation Dose and Imaging
Huazhong University of Science and Technology
2016-2025
Nanjing University of Posts and Telecommunications
2010-2025
Wuhan National Laboratory for Optoelectronics
2021-2025
Hong Kong University of Science and Technology
2003-2024
University of Hong Kong
2003-2024
Guangzhou HKUST Fok Ying Tung Research Institute
2023-2024
PLA Electronic Engineering Institute
2024
Wuhan University
2023-2024
Renmin Hospital of Wuhan University
2023-2024
Shandong University of Science and Technology
2024
Automated methods for prostate cancer (PCa) diagnosis in multi-parametric magnetic resonance imaging (MP-MRIs) are critical alleviating requirements interpretation of radiographs while helping to improve diagnostic accuracy (Artan et al 2010 IEEE Trans. Image Process. 19 2444-55, Litjens 2014 Med. Imaging 33 1083-92, Liu 2013 SPIE Medical (International Society Optics and Photonics) p 86701G, Moradi 2012 J. Magn. Reson. 35 1403-13, Niaf 23 979-91, Phys. Biol. 57 3833, Peng 2013a 86701H,...
Automated methods for detecting clinically significant (CS) prostate cancer (PCa) in multi-parameter magnetic resonance images (mp-MRI) are of high demand. Existing typically employ several separate steps, each which is optimized individually without considering the error tolerance other steps. As a result, they could either involve unnecessary computational cost or suffer from errors accumulated over In this paper, we present an automated CS PCa detection system, where all steps jointly...
Privacy concerns make it infeasible to construct a large medical image dataset by fusing small ones from different sources/institutions. Therefore, federated learning (FL) becomes promising technique learn multi-source decentralized data with privacy preservation. However, the cross-client variation problem in would be bottleneck practice. In this paper, we propose variation-aware (VAFL) framework, where variations among clients are minimized transforming images of all onto common space. We...
Recurrent Neural Networks have long been the dominating choice for sequence modeling. However, it severely suffers from two issues: impotent in capturing very long-term dependencies and unable to parallelize sequential computation procedure. Therefore, many non-recurrent models that are built on convolution attention operations proposed recently. Notably, with multi-head such as Transformer demonstrated extreme effectiveness a variety of modeling tasks. Despite their success, however, these...
As a core task and an important link in the fields of natural language understanding information retrieval, extraction (IE) can structure semanticize unstructured multi-modal information. In recent years, deep learning (DL) has attracted considerable research attention to IE tasks. Deep learning-based entity relation techniques have gradually surpassed traditional feature- kernel-function-based methods terms depth feature model accuracy. this paper, we explain basic concepts DL, primarily...
In this paper, we propose a bi-modality medical image synthesis approach based on sequential generative adversarial network (GAN) and semi-supervised learning. Our consists of two modules that synthesize images the modalities in order. A method for measuring complexity is proposed to automatically determine order our GAN. Images modality with lower are synthesized first, counterparts higher generated later. GAN trained end-to-end manner. supervised training, joint distribution learned from...
Registration of brain MRI images requires to solve a deformation field, which is extremely difficult in aligning intricate tissues, e.g., subcortical nuclei, etc. Existing efforts resort decomposing the target field into intermediate sub-fields with either tiny motions, i.e., progressive registration stage by stage, or lower resolutions, coarse-to-fine estimation full-size field. In this paper, we argue that those are not mutually exclusive, and propose unified framework for robust both...
•Radiomics can improve and refine the pre-operative diagnosis of part-solid nodules.•Radiomic signatures combined with CT features significantly help in differential mixed attenuation ground glass nodules aid differentiation IAs from MIAs.•The quantitative nomogram prediction model based on radiomic score shape could be a step towards precision medicine provide critical information for clinical decision making guiding further management. AIMA was developed to predict histological subtypes...
Multispectral object detection is a crucial technology in remote sensing image processing, particularly low-light environments. Most current methods extract features at single scale, resulting the fusion of invalid and failure to detect small objects. To address these issues, we propose multispectral network based on multilevel feature dual modulation (GMD-YOLO). Firstly, novel dual-channel CSPDarknet53 used deep from visible-infrared images. This incorporates Ghost module, which generates...
Vascular tree disentanglement and vessel type classification are two crucial steps of the graph-based method for retinal artery-vein (A/V) separation. Existing approaches treat them as independent tasks mostly rely on ad hoc rules (e.g. change directions) hand-crafted features color, thickness) to handle respectively. However, we argue that highly correlated should be handled jointly since knowing A/V can unravel those entangled vascular trees, which in turn helps infer types connected...
Inflammatory responses by kidney mesangial cells play a critical role in the glomerulonephritis. The anti-inflammatory potential of nineteen mono-, di- and polyhydroxylated flavones including fisetin, quercetin, morin, tricetin, gossypetin, apigenin myricetin were investigated on rat with lipopolysaccharide (LPS) as inflammatory stimuli. 6-Hydroxyflavone 4′,6-dihydroxyflavone exhibited high activity IC50 range 2.0 μM, much better inhibition comparison to well-studied flavones. Interestingly,...
Political scientists have conducted extensive research on the factors influencing political participation, but empirical analyses examining them from perspective of social fairness perceptions are not common. Using large-scale data Chinese General Social Survey (CGSS), this study explores intermediate mechanisms in influential relationship between capital and farmers' diversified participation based structural equation modeling (SEM). The results show that positive behavior is indirectly...