Format

Send to

Choose Destination
Med Image Anal. 2014 Oct;18(7):1157-68. doi: 10.1016/j.media.2014.06.011. Epub 2014 Jul 11.

In-vivo segmentation and quantification of coronary lesions by optical coherence tomography images for a lesion type definition and stenosis grading.

Author information

1
Scuola Superiore Sant'Anna, Heart Hospital, Massa, Italy; Fondazione CNR/Regione Toscana Gabriele Monasterio FTGM, Heart Hospital, Adult Cardiology Unit, Massa, Italy. Electronic address: s.celi@sssup.it.
2
Fondazione CNR/Regione Toscana Gabriele Monasterio FTGM, Heart Hospital, Adult Cardiology Unit, Massa, Italy.

Abstract

Optical coherence tomography (OCT) is a catheter-based medical imaging technique that produces cross-sectional images of blood vessels. This technique is particularly useful for studying coronary atherosclerosis. In this paper, we present a new framework that allows a segmentation and quantification of OCT images of coronary arteries to define the plaque type and stenosis grading. These analyses are usually carried out on-line on the OCT-workstation where measuring is mainly operator-dependent and mouse-based. The aim of this program is to simplify and improve the processing of OCT images for morphometric investigations and to present a fast procedure to obtain 3D geometrical models that can also be used for external purposes such as for finite element simulations. The main phases of our toolbox are the lumen segmentation and the identification of the main tissues in the artery wall. We validated the proposed method with identification and segmentation manually performed by expert OCT readers. The method was evaluated on ten datasets from clinical routine and the validation was performed on 210 images randomly extracted from the pullbacks. Our results show that automated segmentation of the vessel and of the tissue components are possible off-line with a precision that is comparable to manual segmentation for the tissue component and to the proprietary-OCT-console for the lumen segmentation. Several OCT sections have been processed to provide clinical outcome.

KEYWORDS:

3D reconstruction; Coronary arteries; Optical coherence tomography; Plaque morphology; Vessel segmentation

PMID:
25077844
DOI:
10.1016/j.media.2014.06.011
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center