Journal article
Segmentation of intravascular ultrasound images: a machine learning approach mimicking human vision
International Congress series, Vol.1268(C), pp.1045-1049
2004
DOI: 10.1016/j.ics.2004.03.252
Abstract
This paper describes an approach for fully automated segmentation of intravascular ultrasound (IVUS) images that mimics the procedure performed by human experts. The approach requires no manual initialization or interaction, and is integrated into our previously developed system for reconstruction of coronary arteries by data fusion of IVUS and X-ray angiography. The fusion system is used in an ongoing clinical study assessing the relationships among vessel curvature, plaque thickness, and hemodynamic shear stress. The improved automatic segmentation of IVUS images has resulted in a substantial decrease in the time needed for analysis, thus allowing for an increase in the size and scope of the clinical study. Preliminary results from 10 in vivo cases are presented.
Details
- Title: Subtitle
- Segmentation of intravascular ultrasound images: a machine learning approach mimicking human vision
- Creators
- Mark E Olszewski - University of IowaAndreas Wahle - University of IowaSteven C Mitchell - University of IowaMilan Sonka - University of Iowa
- Resource Type
- Journal article
- Publication Details
- International Congress series, Vol.1268(C), pp.1045-1049
- DOI
- 10.1016/j.ics.2004.03.252
- ISSN
- 0531-5131
- eISSN
- 1873-6157
- Publisher
- Elsevier B.V
- Language
- English
- Date published
- 2004
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Radiation Oncology; The Iowa Institute for Biomedical Imaging; Fraternal Order of Eagles Diabetes Research Center; Injury Prevention Research Center; Ophthalmology and Visual Sciences
- Record Identifier
- 9984186591502771
Metrics
18 Record Views