A5100714747- Advanced Image Processing Techniques
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Advanced Neuroimaging Techniques and Applications
- Advanced MRI Techniques and Applications
- Image and Signal Denoising Methods
- Advanced Image and Video Retrieval Techniques
- COVID-19 diagnosis using AI
- Visual Attention and Saliency Detection
- Advanced Vision and Imaging
- Advanced Neural Network Applications
- MRI in cancer diagnosis
- Image Processing Techniques and Applications
- Image Enhancement Techniques
- RNA and protein synthesis mechanisms
- Medical Imaging and Analysis
- Advanced Image Fusion Techniques
- Color Science and Applications
- Industrial Vision Systems and Defect Detection
- Machine Learning in Bioinformatics
- Bioinformatics and Genomic Networks
- Grouting, Rheology, and Soil Mechanics
- Face recognition and analysis
- Image and Video Quality Assessment
- Image Retrieval and Classification Techniques
Heilongjiang University of Science and Technology
2009-2024
Heilongjiang University
2006-2024
China University of Mining and Technology
2022
Shanghai Medical Information Center
2020
Harbin Institute of Technology
2006-2016
Harbin Medical University
2014
An image retrieval system is presented, which used HSV color space and wavelet transform approach for feature extraction. Firstly, we quantified the in non-equal intervals, then constructed one dimension vector represented feature. Similarly, work of texture extraction obtained by using wavelet. Finally, combine based on transform. A method multi features provided. The experiments indicated that visual were sensitive different type images. opted to rich with simple variety. Texture complex...
The rapidly emerging field of deep learning-based computational pathology has shown promising results in utilizing whole slide images (WSIs) to objectively prognosticate cancer patients. However, most prognostic methods are currently limited either histopathology or genomics alone, which inevitably reduces their potential accurately predict patient prognosis. Whereas integrating WSIs and genomic features presents three main challenges: (1) the enormous heterogeneity gigapixel can reach sizes...
Cancer is widely recognized as the primary cause of mortality worldwide, and pathology analysis plays a pivotal role in achieving accurate cancer diagnosis. The intricate representation features histopathological images encompasses abundant information crucial for disease diagnosis, regarding cell appearance, tumor microenvironment, geometric characteristics. However, recent deep learning methods have not adequately exploited pathological image classification due to absence effective...
Identifying dysregulated pathways from high-throughput experimental data in order to infer underlying biological insights is an important task. Current pathway-identification methods focus on single isolation; however, consideration of crosstalk between could improve our understanding alterations states. We propose a novel method pathway analysis based global influence (PAGI) identify pathways, by considering both within-pathway effects and pathways. constructed gene–gene network the...
High-throughput metabolomics technology, such as gas chromatography mass spectrometry, allows the analysis of hundreds metabolites. Understanding that these metabolites dominate study condition from biological pathway perspective is still a significant challenge. Pathway identification an invaluable aid to address this issue and, thus, urgently needed. In study, we developed network-based metabolite method, MPINet, which considers global importance and unique character metabolomic profile....
Aberrant metabolism is one of the main driving forces in initiation and development ESCC. Both genes metabolites play important roles metabolic pathways. Integrative pathway analysis both will thus help to interpret underlying biological phenomena. Here, we performed integrative gene metabolite profiles by analyzing six expression seven Multiple known novel subpathways associated with ESCC, such as 'beta-Alanine metabolism', were identified via cooperative use differential genes,...
Identifying enhancers is a critical task in bioinformatics due to their primary role regulating gene expression. For this reason, various computational algorithms devoted enhancer identification have been put forward over the years. More features are extracted from single DNA sequences boost performance. Nevertheless, structural information neglected, which an essential factor affecting binding preferences of transcription factors regulatory elements like enhancers. Here, we propose SENIES,...
Automatic retinal vessel segmentation in fundus image can assist effective and efficient diagnosis of retina disease. Microstructure estimation capillaries is a prolonged challenging issue. To tackle this problem, we propose attention-aware multi-scale fusion network (AMF-Net). Our with dense convolutions to perceive microscopic capillaries. Additionally, features are extracted fused adaptive weights by channel attention module improve the performance. Finally, spatial introduced position...
Gaussian Mixture Model (GMM) is a popular tool for density estimation. The parameters of the GMM are estimated based on Maximum Likelihood principle (MLP) in almost all recognition system. However, number mixtures used model important determining model's effectiveness; general problem mixture modeling difficult when components unknown. This paper presents currency using Structural Risk Minimization (SRM). By selecting proper with SRM, system can overcome demerit by selected artificially. A...
Continual increases in IC-chip complexity and performance are placing demands on the density functionality of package I/Os. Therefore, various SMT interconnection techniques being developed to satisfy this need, including ball-grid-array (BGA). BGA has been used a production PCB (printed circuit board), because their excellent characters such as high lead pin pitch that can achieve 400 I/Os per square inch. This paper deals with detection defects at solder joints PC boards by using X-ray...
Machine vision has been widely used in various industrial productions. However, the study for solder joints detection is not enough. This paper presents a method based on surface recovery. For single gray-scale image, using shape-from-shading (SFS) technology, of recovered. According to shape distribution, quality discriminated. In order improve accuracy recovery real images, hybrid illumination model introduced and reflection-component estimation simulated annealing algorithm designed. Then...
In computer vision, low-light image enhancement has always been a challenging task caused by more lower signal to noise ratio. Some methods have proposed enhance the using fully convolution network. Using u-net as backbone, we introduce wavelet transform conduct down-sampling and up-sampling operations. order recover details, perceptual loss used optimize network parameters. Experiments show that our model can get better performance than existing methods. We find effectively improve quality...
The histopathology analysis is of great significance for the diagnosis and prognosis cancers, however, it has challenges due to enormous heterogeneity gigapixel whole slide images (WSIs) intricate representation pathological features. However, recent methods have not adequately exploited geometrical in WSIs which significant disease diagnosis. Therefore, we proposed a novel weakly-supervised framework, Geometry-Aware Transformer (GOAT), urge model pay attention geometric characteristics...
Abstract The segmentation of atrial scars in LGE-MRI images has huge potential value for clinical diagnosis and subsequent treatment. In practice, are usually manually calibrated by experienced experts, which is time-consuming prone to errors. However, automatic also faces difficulties due myocardial scars’ small size variable shape. present study introduces a dual branch network, incorporating edge attention, deep supervision strategy. Edge attention introduced fully utilize the spatial...