Ruoxiu Xiao

ORCID: 0000-0003-3959-9797
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Medical Image Segmentation Techniques
  • Cerebrovascular and Carotid Artery Diseases
  • Retinal Imaging and Analysis
  • Advanced X-ray and CT Imaging
  • Advanced MRI Techniques and Applications
  • Orthopaedic implants and arthroplasty
  • Advanced Vision and Imaging
  • Total Knee Arthroplasty Outcomes
  • Optical measurement and interference techniques
  • Acute Ischemic Stroke Management
  • Anatomy and Medical Technology
  • Advanced Image Processing Techniques
  • Digital Imaging in Medicine

Tsinghua University
2015-2018

University of Science and Technology Beijing
2018

Beijing Institute of Technology
2013-2016

As the uneven distribution of contrast agents and perspective projection principle X-ray, vasculatures in angiographic image are with low generally superposed other organic tissues; therefore, it is very difficult to identify vasculature quantitatively estimate blood flow directly from images. In this paper, we propose a fully automatic algorithm named adaptive geometrical vessel tracking (AGVT) for coronary artery identification X-ray angiograms. Initially, ridge enhancement (RE) obtained...

10.1155/2013/796342 article EN cc-by Computational and Mathematical Methods in Medicine 2013-01-01

Seed point is prerequired condition for tracking based method extracting centerline or vascular structures from the angiogram. In this paper, a novel seed detection coronary artery segmentation proposed. Vessels on image are first enhanced according to distribution of Hessian eigenvalue in multiscale space; consequently, centerlines tubular vessels also enhanced. Ridge extracted as candidate point, which then refined its mathematical definition. The theoretical feasibility proven. Finally,...

10.1155/2015/502573 article EN cc-by Computational and Mathematical Methods in Medicine 2015-01-01

In this paper, a novel method is developed for the 3-D reconstruction of coronary artery trees from two angiographic images. Due to rotational distortion and unpredictable motion imaging chain, commonly used pinhole camera model cannot obtain desirable vascular trees. Curvature Scale Space (CSS) approach find correspondence by choosing similar views between various obtained technique in one cardiac cycle. Therefore, study curve matching epipolar constraint are integrated calculate...

10.1109/icccnt.2013.6726682 article EN 2013 Fourth International Conference on Computing, Communications and Networking Technologies (ICCCNT) 2013-07-01
Coming Soon ...