A5071335209- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Medical Image Segmentation Techniques
- Retinal Imaging and Analysis
- Digital Imaging for Blood Diseases
- Gene expression and cancer classification
- Advanced Image and Video Retrieval Techniques
- Retinal and Optic Conditions
- Image Retrieval and Classification Techniques
- Advanced Neural Network Applications
- Advanced Computational Techniques and Applications
- Colorectal Cancer Screening and Detection
- Chaos-based Image/Signal Encryption
- Advanced Decision-Making Techniques
- Medical Imaging and Analysis
- Glaucoma and retinal disorders
- Computer Graphics and Visualization Techniques
- Indoor and Outdoor Localization Technologies
- Rough Sets and Fuzzy Logic
- Cognitive Science and Education Research
- Brain Tumor Detection and Classification
- 3D Shape Modeling and Analysis
- Advanced Malware Detection Techniques
- Cell Image Analysis Techniques
- Traditional Chinese Medicine Studies
Guilin University of Electronic Technology
2012-2025
Guilin University
2009
In the field of gland segmentation in histopathology, deep-learning methods have made significant progress. However, most existing not only require a large amount high-quality annotated data but also tend to confuse internal with background. To address this challenge, we propose new semi-supervised method named DCCL-Seg for segmentation, which follows teacher-student framework. Our approach can be divided into steps. First, design contrastive learning module improve ability student model's...
Image-level weakly supervised semantic segmentation (WSSS) reduces the dependence on high-quality data annotation, which plays a crucial role in computational pathology. Benefit from ability to localize objects with only binary labels, Class Activation Map (CAM) is widely used method initial pseudo masks. However, due low contrast among different tissues histopathological images, most existing CAM-based methods perform poorly gland segmentation. We retrospect this process and find that class...
Focused on the problem of low accuracy floor positioning using WiFi fingerprints, a method based fusion stack autoencoder and multiple models was proposed. In process locating through fingerprint, fingerprint database usually contains information hundreds APs, resulting in "dimensional disaster". order to solve this problem, for collected data, feature selection is first performed by random forest algorithm, AP points that have little effect result are found deleted from database; secondly,...
Man-in-the-middle attack on the BB84 protocol is presented. The man in middle can intercept message from sender and send his to receiver every step of protocol. be analogously used other quantum key distribution schemes. Some methods are suggested defense against by discovering or handicapping attack, security analyzed.
Retinal fundus image segmentation based on deep learning is an important method for auxiliary diagnosis of ophthalmic diseases. Due to the scale difference blood vessels and imbalance between foreground background pixels in image, network will inevitably ignore thin when downsampling feature learning. For problem, this article aims tackle its limitation from two perspectives: changing supervised approach adapting Correspondingly, a dilation-supervised adaptive dimensional attention mechanism...
Accurate and automated gland segmentation on pathological images can assist pathologists in diagnosing the malignancy of colorectal adenocarcinoma. However, due to various shapes, severe deformation malignant glands, overlapping adhesions between glands. Gland has always been very challenging. To address these problems, we propose a DEA model. This model consists two branches: backbone encoding decoding network local semantic extraction network. The extracts advanced Semantic features, uses...
It is pointed out that the reliability of information largely ignored and not absolutely reliable widely existent. The significance studying explicated. consideration in Shannon theory analyzed, relation between uncertainty illustrated. clarified related to multi-uncertainty event, is, probability event.
Ray-casting algorithm is an important volume rendering algorithm, which widely used in medical image processing. Aiming to address the shortcomings of current ray-casting algorithms 3D reconstruction images, such as slow speed and low sampling efficiency, improved based on dynamic adaptive proposed. By using central difference gradient method, corresponding interval obtained dynamically according different points. Meanwhile, a new operator proposed color value opacity changes before after...
By borrowing ideas from a cryptographic algorithm of low key authentic degree, novel steganographic method based on keyword shift is presented. The master the to sensitive keywords in text. conditions guarantee reversibility are analyzed and found out, serviceability some situations pointed out. strong points weak analyzed.
Deep learning-based vessel segmentation tasks serve as important auxiliary tools for disease diagnosis. However, the region connecting foreground and background, which is named edge-buffer in this paper, suffers from noisy labels a lack of discriminative features due to low contrast limitations imaging devices. To address these limitations, we propose self-correcting supervision revamp region. Furthermore, introduce concept treating independently leveraging designed contrastive learning...
The lack of developing platforms and the incompatibilities protocols result in difficulties rapidly P2P applications, JXTA proposed by SUN corporation is an opensource platform to solve these problems, but it still needs be improved. characteristics between mobile agent are analyzed this paper. A novel architecture based on proposed. new resource expressing method referring Dublin Core metadata corresponding message format designed, a searching algorithm designed. At last prototype system...