A5007211501- Retinal Imaging and Analysis
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Glaucoma and retinal disorders
- Advanced Neural Network Applications
- EEG and Brain-Computer Interfaces
- Retinal Diseases and Treatments
- Brain Tumor Detection and Classification
- Multimodal Machine Learning Applications
- MRI in cancer diagnosis
- Image Processing Techniques and Applications
- Digital Imaging for Blood Diseases
- Smart Agriculture and AI
- Medical Imaging and Analysis
- Digital Radiography and Breast Imaging
- Functional Brain Connectivity Studies
- Advanced Vision and Imaging
- COVID-19 diagnosis using AI
- Medical Imaging Techniques and Applications
- Image and Signal Denoising Methods
- Advanced Image Processing Techniques
- Neural and Behavioral Psychology Studies
Zhengzhou University
2025
Henan University of Technology
2025
Shandong Normal University
2015-2024
Beijing Academy of Artificial Intelligence
2023-2024
Nanchang Hangkong University
2024
Shanghai Artificial Intelligence Laboratory
2023-2024
South China University of Technology
2023-2024
Chongqing Public Health Medical Center
2023
Chongqing Medical University
2023
State Grid Corporation of China (China)
2022-2023
A variant of the popular nonparametric nonuniform intensity normalization (N3) algorithm is proposed for bias field correction. Given superb performance N3 and its public availability, it has been subject several evaluation studies. These studies have demonstrated importance certain parameters associated with B-spline least-squares fitting. We propose substitution a recently developed fast robust approximation routine modified hierarchical optimization scheme improved correction over...
Automated breast cancer multi-classification from histopathological images plays a key role in computer-aided diagnosis or prognosis. Breast is to identify subordinate classes of (Ductal carcinoma, Fibroadenoma, Lobular etc.). However, faces two main challenges from: (1) the great difficulties methods contrasting with classification binary (benign and malignant), (2) subtle differences multiple due broad variability high-resolution image appearances, high coherency cancerous cells, extensive...
Feature matching, which refers to establishing the correspondence of regions between two images (usually voxel features), is a crucial prerequisite feature-based registration. For deformable image registration tasks, traditional methods typically use an iterative matching strategy for interest region where feature selection and are explicit, but specific schemes often useful in solving application-specific problems require several minutes each In past few years, feasibility learning-based...
The amount of fibroglandular tissue content in the breast as estimated mammographically, commonly referred to percent density (PD%), is one most significant risk factors for developing cancer. Approaches quantify focus on either semiautomated methods or visual assessment, both which are highly subjective. Furthermore, studies published date investigating computer-aided assessment PD% have been performed using digitized screen-film mammograms, while digital mammography increasingly replacing...
We cast some new insights into solving the digital matting problem by treating it as a semi-supervised learning task in machine learning. A local based approach and global are then produced, to fit better scribble trimap matting, respectively. Our approaches easy implement because only simple matrix operations needed. They also extremely accurate they can efficiently handle nonlinear color distributions incorporating kernel trick, that beyond ability of many previous works. outperform recent...
Purpose: Mammographic percent density (PD%) is known to be a strong risk factor for breast cancer. Recent studies also suggest that parenchymal texture features, which are more granular descriptors of the pattern, can provide additional information about cancer risk. To date, most have measured mammographic within selected regions interest (ROIs) in breast, cannot adequately capture complexity pattern throughout whole breast. better characterize patterns tissue, authors developed fully...
Surface detection of small defects plays a vital role in manufacturing and has attracted broad interest. It remains challenging primarily due to the size defect relative large surface rare occurrence defects. To address this problem, article we propose novel machine vision approach for automatically identifying tiny flaws that may appear single image. First, presented exaggeration produces both flawless image corresponding exaggerated version by taking variations as regularization terms....
Hippocampal morphological change is one of the main hallmarks Alzheimer's disease (AD). However, whether hippocampal radiomic features are robust as predictors progression from mild cognitive impairment (MCI) to AD dementia and these provide any neurobiological foundation remains unclear. The primary aim this study was verify can serve magnetic resonance imaging (MRI) markers for AD. Multivariate classifier-based support vector machine (SVM) analysis provided individual-level predictions...
We propose a new multiscale rotation-invariant convolutional neural network (MRCNN) model for classifying various lung tissue types on high-resolution computed tomography. MRCNN employs Gabor-local binary pattern that introduces good property in image analysis-invariance to scales and rotations. In addition, we offer an approach deal with the problems caused by imbalanced number of samples between different classes most existing works, accomplished changing overlapping size adjacent patches....
Logo detection has been gaining considerable attention because of its wide range applications in the multimedia field, such as copyright infringement detection, brand visibility monitoring, and product management on social media. In this article, we introduce LogoDet-3K, largest logo dataset with full annotation, which 3,000 categories, about 200,000 manually annotated objects, 158,652 images. LogoDet-3K creates a more challenging benchmark for higher comprehensive coverage wider variety...
Electrochemical production of H2O2 is a cost-effective and environmentally friendly alternative to the anthraquinone-based processes. Metal-doped carbon-based catalysts are commonly used for 2-electron oxygen reduction reaction (2e-ORR) due their high selectivity. However, exact roles metals carbon defects on ORR remain unclear. Herein, by varying Co loading in pyrolysis precursor, Co-N/O-C catalyst with Faradaic efficiency greater than 90% alkaline electrolyte was obtained. Detailed studies...
In this paper, we propose a method for robustly determining the vignetting function given only single image. Our is designed to handle both textured and untextured regions in order maximize use of available information. To extract information from an image, present adaptations segmentation techniques that locate image with reliable data estimation. Within each region, our capitalizes on frequency characteristics physical properties distinguish it other sources intensity variation. Rejection...
Determining banana’s ripening stages is becoming an essential requirement for standardizing the quality of commercial bananas. In this paper, we propose a novel convolutional neural network architecture which designed specifically fine-grained classification stages. It learns set image features based on data-driven mechanism and offers deep indicator stage. The resulted can help to differentiate subtle differences among subordinate classes bananas in state. Experimental results from 17,312...