A5100323843- Medical Image Segmentation Techniques
- Image and Object Detection Techniques
- Image Retrieval and Classification Techniques
- Radiomics and Machine Learning in Medical Imaging
- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Medical Imaging Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Robotics and Automated Systems
- Virtual Reality Applications and Impacts
- Advanced Measurement and Detection Methods
- Robotic Path Planning Algorithms
- Augmented Reality Applications
- Advanced Image and Video Retrieval Techniques
- Optical measurement and interference techniques
- Neuroscience and Neural Engineering
- Advanced Radiotherapy Techniques
- Interactive and Immersive Displays
- Image Processing Techniques and Applications
- Advanced X-ray and CT Imaging
- Luminescence and Fluorescent Materials
- Image Processing and 3D Reconstruction
- Advanced Sensor and Energy Harvesting Materials
- Medical Imaging and Analysis
Xi'an University of Technology
2021-2024
University Hospital Magdeburg
2019-2021
BOE Technology Group (China)
2018
Guangxi University
2013-2014
Beihang University
2014
Central South University
2013
Air Force Engineering University
2010
As a challenging task in computer vision, instance segmentation has attracted extensive attention recent years. Able to obtain very rich and refined object information, this technology shows important application value many fields, such as intelligent driving, medical health, remote sensing detection. Instance should not only identify the positions of objects but also accurately mark boundary any single instance, which can be defined solving detection semantic at same time. Our study gives...
Traditional watershed segmentation is sensitive to noise and can leads serious over-segmentation. In order overcome the shortcomings of traditional segmentation, this paper presented an improved image method. Firstly, morphological opening/closing reconstruction filter applied remove noise. Secondly, multi-scale structure elements are used calculate gradient. Furthermore, gradient modified by viscous operators which most irregular local minimums. After standard transform, region merging...
In recent years, image-guided thermal ablations have become a considerable treatment method for cancer patients, including support through navigational systems. One of the most critical challenges in these systems is registration between intraoperative images and preoperative volume. The motion secondary to inspiration makes even more difficult. this work, we propose coarse-fine fast patient technique solve problem compensation. contrast other state-of-the-art methods, focus on improving...
2D-US to 3D-CT/MR registration is a crucial module during minimally invasive ultrasound-guided liver tumor ablations. Many modern methods still require manual or semi-automatic slice pose initialization due insufficient robustness of automatic methods. The state-of-the-art regression networks do not work well for 2D US 3D CT/MR because the tremendous inter-patient variability anatomy. To address this unsolved problem, we propose deep learning network pipeline which - instead starts with...
With the variety of medical imaging equipment’s application in process,medical image registration becomes particularly important field processing,which has clinical diagnostic and therapeutic value. This article describes matrix conversion method rigid model, basic concepts principles mutual information algorithm ,the idea genetic algorithms algorithm’s flow , improved practice. The two CT brain bones images uses as a similarity measure, search strategy matlab programming environment. Using...
Most of the current thresholding algorithms for gray images conduct segmentation on 2-dimensional histogram, resulting in low precisions details and weak immunities noises. A novel algorithm based 3-dimensional Renyi's entropy is proposed this paper. The new dimension added weighted neighbor median grayscale, which has shown a high effectiveness noise removal detail reservation, thereby improved quality. experiments different indicated that our increased precision raised immunity compared...
Object : To optimize the rigidity registration algorithm between X-ray fluoroscopy and CT, improve accuracy of registration. Method By changing transmission parameters ray tracing, it can obtain original DRR images float image for In trials, uses ant colony as optimized search strategy Mutual information similarity measure. Result: improved compared to classic Powell about 10% 20%, achieved good results. Conclusion Ant Colony Algorithm optimization effectively solve local minima problem in...
Abstract Background and Objective: Ultrasound (US) devices are often used in percutanous interventions. Due to their low image quality, the US slices aligned with pre-operative Computed Tomography/Magnetic Resonance Imaging (CT/MRI) images enable better visibilities of anatomies during intervention. This work aims at improving deep learning one shot registration by using less loops through networks.Methods: We propose two cascade networks which aim accuracy loops. The...
Augmented Reality (AR) is a technique that uses computer generated additional information to augment or expand the real world view, and optical device one of most core components. This paper introduces our research about AR system applied in lightweight waveguide with low thickness light weight structure, discussed stray phenomenon, as well glasses can be implemented only FPGA chip distortion correction.
To accurately extract the gum line of teeth from three-dimensional dental model, a new method based on traditional Dijkstra shortest path algorithm with Sigmoid curvature transformation is proposed in this paper. The each vertex model transformed sigmoid function, and edge assigned corresponding weight, then representing calculated. Also, queue optimization storage structure strategy used to make implementation efficient. experimental results show that tooth segmentation can boundary accurately.