A5013521351- Retinal Imaging and Analysis
- Retinal Diseases and Treatments
- Glaucoma and retinal disorders
- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- Retinal and Optic Conditions
- Digital Imaging for Blood Diseases
- Brain Tumor Detection and Classification
- Mycorrhizal Fungi and Plant Interactions
- Video Analysis and Summarization
- Generative Adversarial Networks and Image Synthesis
- AI in cancer detection
- Combustion and Detonation Processes
- Wind and Air Flow Studies
- Ecosystem dynamics and resilience
- Nematode management and characterization studies
- Forest Insect Ecology and Management
- COVID-19 diagnosis using AI
- Reproductive Biology and Fertility
- Greenhouse Technology and Climate Control
- Advanced Image Processing Techniques
- Advanced Thermodynamic Systems and Engines
- Molecular Biology Techniques and Applications
- Sustainability and Ecological Systems Analysis
Beihang University
2018-2024
Yantai Nanshan University
2024
Beijing Advanced Sciences and Innovation Center
2019-2023
Hefei Institute of Technology Innovation
2019-2023
Anhui Medical University
2019-2023
Deep learning methods have been successfully applied in medical image classification, segmentation and detection tasks. The U-Net architecture has widely for these In this paper, we propose a variant improved vessel retinal fundus images. Firstly, design minimal (Mi-UNet) architecture, which drastically reduces the parameter count to 0.07M compared 31.03M conventional U-Net. Moreover, based on Mi-UNet, Salient (S-UNet), bridge-style with saliency mechanism only 0.21M parameters. S-UNet uses...
Semantic segmentation of medical images provides an important cornerstone for subsequent tasks image analysis and understanding. With rapid advancements in deep learning methods, conventional U-Net networks have been applied many fields. Based on exploratory experiments, features at multiple scales found to be great importance the images. In this paper, we propose a scale-attention network (SA-Net), which extracts different residual module uses attention enforce capability. SA-Net can better...
Retinal blood vessel morphological abnormalities are generally associated with cardiovascular, cerebrovascular, and systemic diseases, automatic artery/vein (A/V) classification is particularly important for medical image analysis clinical decision making. However, the current method still has some limitations in A/V classification, especially edge end error problems caused by single scale blurred boundary of A/V. To alleviate these problems, this work, we propose a vessel-constraint network...
Automatic classification of retinal arteries and veins plays an important role in assisting clinicians to diagnosis cardiovascular eye-related diseases. However, due the high degree anatomical variation across population, presence inconsistent labels by subjective judgment annotators available training data, most existing methods generally suffer from blood vessel discontinuity arteriovenous confusion, artery/vein (A/V) task still faces great challenges. In this work, we propose a...
The quality of a blastocyst directly determines the embryo's implantation potential, thus making it essential to objectively and accurately identify morphology. In this work, we propose an automatic framework named I2CNet perform segmentation task in human embryo images. contains two components: IntrA-Class Context Module (IACCM) InteR-Class (IRCCM). IACCM aggregates representations specific areas sharing same category for each pixel, where categorized regions are learned under supervision...
Early and accurate diagnosis of glaucoma is critical for avoiding human vision deterioration preventing blindness. A deep‐neural‐network model has been developed the based on Heidelberg retina tomography (HRT), called “Seeking Common Features Reserving Differences Net” (SCRD‐Net) to make full use HRT data. In this work, proposed SCRD‐Net achieved an area under curve (AUC) 94.0%. For two image modalities, sensitivities were 91.2% 78.3% at specificities 0.85 0.95, respectively. These results...
The segmentation of blastocyst components is vital in assessing embryo quality, as implantation potential closely relates to morphological characteristics. Despite this significance, automated encounters challenges like poor contrast, noise, and indistinct boundaries among organizational structures. In response, we present a novel transformer architecture called BTFormer for segmentation. Our approach integrates an axial-free attention mechanism with reduced computational demands,...
In the above article <xref ref-type="bibr" rid="ref1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[1]</xref> , we discovered some errors mainly involving network parameters, test results of TONGREN training data on DRIVE and CHASE_DB1 in ref-type="table" rid="table4" xmlns:xlink="http://www.w3.org/1999/xlink">Table 4</xref> reference [7], minor writing errors. These do not affect final conclusions published paper. The details corrections are as follows: