A5038053132- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- Retinal Imaging and Analysis
- Advanced X-ray and CT Imaging
- Advanced Vision and Imaging
- Image and Object Detection Techniques
- Medical Imaging Techniques and Applications
- Advanced Image Processing Techniques
- COVID-19 diagnosis using AI
- Radiation Dose and Imaging
- Coronary Interventions and Diagnostics
- Anatomy and Medical Technology
- Image Enhancement Techniques
- Radiomics and Machine Learning in Medical Imaging
- Optical measurement and interference techniques
- Infrastructure Maintenance and Monitoring
- Advanced MRI Techniques and Applications
- 3D Surveying and Cultural Heritage
- Brain Tumor Detection and Classification
- COVID-19 epidemiological studies
- Anomaly Detection Techniques and Applications
Tsinghua University
2020-2024
Abstract Background Coronary artery angiography is an indispensable assistive technique for cardiac interventional surgery. Segmentation and extraction of blood vessels from coronary angiographic images or videos are very essential prerequisites physicians to locate, assess diagnose the plaques stenosis in vessels. Methods This article proposes a novel segmentation framework that combines three–dimensional (3D) convolutional input layer two–dimensional (2D) network. Instead single image...
Abstract Background Coronary heart disease is one of the diseases with highest mortality rate. Due to important position cardiovascular prevention and diagnosis in medical field, segmentation images has gradually become a research hotspot. How segment accurate blood vessels from coronary angiography videos assist doctors making analysis goal our research. Method Based on U-net architecture, we use context-based convolutional network for capturing more information vessel video. The proposed...
Abstract Background: With the current worldwide spreading of coronary virus (COVID-19) pandemic, accurately predicting rate spread has become an urgent need. Methods: In this article we propose a universal COVID-19 prediction model that is independent country-specific factors in paper. By analyzing pandemic data China, combined advantages Gaussian function with chi-square distribution function, to render innovative mathematical named H-Gaussian five parameters be learned, and solved by...
Swin Transformer is an important work among all the attempts to reduce computational complexity of Transformers while maintaining its excellent performance in computer vision. Window-based patch self-attention can use local connectivity image features, and shifted window-based enables communication information between different patches entire scope. Through in-depth research on effects sizes windows efficiency, this article proposes a Dual-Scale with double-sized window attention method. The...
Cardiac coronary angiography is a major technique that assists physicians during interventional heart surgery. Under X-ray irradiation, the physician injects contrast agent through catheter and determines arteries’ state in real time. However, to obtain more accurate of arteries, need increase frequency intensity exposure, which will inevitably potential for harm both patient surgeon. In work reported here, we use advanced deep learning algorithms find method frame interpolation videos...
The reconstruction of three-dimensional models coronary arteries is great significance for the localization, evaluation and diagnosis stenosis plaque in arteries, as well assisted navigation interventional surgery. In clinical practice, physicians use a few angles angiography to capture arterial images, so it practical value perform 3D directly from images. However, this very difficult computer vision task due complex shape blood vessels, lack data set key point labeling. With rise deep...
Cardiac coronary angiography is a major technology to assist doctors during cardiac interventional surgeries. Under the exposure of X-ray radiation, inject contrast agents through catheters determine position and status vessels in real time. To get video with high frame rate, doctor needs increase frequency intensity X-ray. This will inevitably harm both patients surgeons. In this work, we innovatively utilize deep-learning based interpolation algorithm interpolate videos. Moreover,...
This study aims to evaluate the clinical application value of flash spiral mode high-pitch dual source CT in carotid, cardiac and cerebral vessels combined one-stop imaging.A total 100 consecutive patients were given imaging at CT. 27 received DSA examination carotid vessels, 38 digital subtraction angiography (DSA) coronary artery same time. Carotid, was compared with "golden standard", image.The overall satisfaction rate arteries, extracranial segment (CA-E), intracranial (CA-I),...
Coronary angiography is an indispensable assistive technique for cardiac interventional surgery. Segmentation and extraction of blood vessels from coronary videos are very essential prerequisites physicians to locate, assess diagnose the plaques stenosis in vessels. This article proposes a new video segmentation framework that can extract clearest most comprehensive images sequence, thereby helping better observe condition combines 3D convolutional layer spatial--temporal information...