A5059912382- Image Processing Techniques and Applications
- Optical measurement and interference techniques
- Advanced Neural Network Applications
- AI in cancer detection
- Anomaly Detection Techniques and Applications
- Medical Image Segmentation Techniques
- Image and Object Detection Techniques
- Cell Image Analysis Techniques
- Industrial Vision Systems and Defect Detection
- Radiomics and Machine Learning in Medical Imaging
- Digital Media Forensic Detection
- Advanced Measurement and Metrology Techniques
- Optical Systems and Laser Technology
- Face recognition and analysis
- Advanced Measurement and Detection Methods
- Advanced Computational Techniques and Applications
- Image and Signal Denoising Methods
- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Forensic and Genetic Research
- Advanced X-ray Imaging Techniques
- COVID-19 diagnosis using AI
- Sperm and Testicular Function
- Single-cell and spatial transcriptomics
- Gene expression and cancer classification
Yantai University
2025
Xidian University
2025
Chinese Academy of Sciences
2022-2024
University of Notre Dame
2018-2024
Xi'an Jiaotong University
2019-2024
Changchun Institute of Optics, Fine Mechanics and Physics
2022-2024
The University of Texas Southwestern Medical Center
2024
Jilin Agricultural Science and Technology University
2023
Beijing Research Institute of Automation for Machinery Industry (China)
2021-2023
Cancer Research Institute
2020
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
3D image segmentation plays an important role in biomedical analysis. Many 2D and deep learning models have achieved state-of-the-art performance on datasets. Yet, their own strengths weaknesses, by unifying them together, one may be able to achieve more accurate results. In this paper, we propose a new ensemble framework for that combines the merits of models. First, develop fully convolutional network based meta-learner learn how improve results from (base-learners). Then, minimize...
Deep learning has been applied successfully to many biomedical image segmentation tasks. However, due the diversity and complexity of data, manual annotation for training common deep models is very timeconsuming labor-intensive, especially because normally only experts can annotate data well. Human are often involved in a long iterative process annotation, as active type schemes. In this paper, we propose representative (RA), new framework reducing effort segmentation. RA uses unsupervised...
Pathology images contain rich information of cell appearance, microenvironment, and topology features for cancer analysis diagnosis. Among such features, becomes increasingly important in immunotherapy. By analyzing geometric hierarchically structured distribution topology, oncologists can identify densely-packed cancer-relevant communities (CCs) making decisions. Compared to commonly-used pixel-level Convolution Neural Network (CNN) cell-instance-level Graph (GNN) CC are at a higher level...
The increasing demand on electrical power consumption all over the world makes need for stable and reliable grids is indispensable. Meanwhile, grid fault diagnosis based recording data an important technology to ensure normal operation of grid. Despite fact that dozens studies have been put forward detect faults, these still suffer from several downsides, such as fuzzy characteristics complex samples with small inter-class differences large intra-class in different topology structures...
Lithium dendrites are widely acknowledged as the main culprit of degradation performance in various Li-based batteries. Studying mechanism lithium dendrite formation is challenging because high reactivity metal. In this work, a phase field model and situ observation experiments were used to study growth kinetics morphologies terms anisotropy, temperature, potential difference. Subsequently, 2D numerical simulation has been developed illustrate impact microscopic parameters, such electrolyte...
Abstract Cell-cell communication (CCC) is essential to how life forms and functions. However, accurate, high-throughput mapping of expression all genes in one cell affects another made possible only recently through the introduction spatially resolved transcriptomics (SRT) technologies, especially those that achieve single-cell resolution. Nevertheless, substantial challenges remain analyze such highly complex data properly. Here, we introduce a multiple-instance learning framework, Spacia,...
In recent years, deep learning (DL) methods have become powerful tools for biomedical image segmentation. However, high annotation efforts and costs are commonly needed to acquire sufficient training data DL models. To alleviate the burden of manual annotation, in this paper, we propose a new weakly supervised approach segmentation using boxes only annotation. First, develop method combine graph search (GS) generate fine object masks from box which uses compute rough GS then is applied...
Purpose Metal implants in the patient's body can generate severe metal artifacts x‐ray computed tomography (CT) images. These may cover tissues around CT images and even corrupt tissue regions, thus affecting disease diagnosis using these Previous deep learning trace inpainting methods used both valid pixels of uncorrupted areas invalid corrupted to patch (i.e., holes removed metal‐corrupted regions). Such cannot recover fine details well often suffer information mismatch due interference...
In three dimensional profilometry, phase retrieval technique plays a key role in signal processing stage. Fringe images need to be transformed into information obtain the measurement result. this paper, new method based on deep learning is proposed for interferometry. Different from conventional multi-step shift methods, can extracted only single frame of an interferogram by method. Here, task regarded as regression problem and hypercolumns convolutional neural network constructed solve it....
In the pinhole point diffraction interferometer (PPDI), proper alignment between reflection spot of tested component and is critical to obtain accurate interferograms. At present, adjusting for tilt error requires manual manipulation, defocus cannot be corrected. These limitations impede instrumentation process PPDI. To address this issue, proposed mirror system utilizes theory analyze mathematical caused by misalignment mirror's reflected beam pinhole. An based on machine vision was...