Share Email Print

Proceedings Paper

Support vector machine based classification and mapping of atherosclerotic plaques using fluorescence lifetime imaging (Conference Presentation)
Author(s): Hussain Fatakdawala; Dimitris S. Gorpas; Julien Bec; Dinglong M. Ma; Diego R. Yankelevich; John W. Bishop; Laura Marcu
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The progression of atherosclerosis in coronary vessels involves distinct pathological changes in the vessel wall. These changes manifest in the formation of a variety of plaque sub-types. The ability to detect and distinguish these plaques, especially thin-cap fibroatheromas (TCFA) may be relevant for guiding percutaneous coronary intervention as well as investigating new therapeutics. In this work we demonstrate the ability of fluorescence lifetime imaging (FLIm) derived parameters (lifetime values from sub-bands 390/40 nm, 452/45 nm and 542/50 nm respectively) for generating classification maps for identifying eight different atherosclerotic plaque sub-types in ex vivo human coronary vessels. The classification was performed using a support vector machine based classifier that was built from data gathered from sixteen coronary vessels in a previous study. This classifier was validated in the current study using an independent set of FLIm data acquired from four additional coronary vessels with a new rotational FLIm system. Classification maps were compared to co-registered histological data. Results show that the classification maps allow identification of the eight different plaque sub-types despite the fact that new data was gathered with a different FLIm system. Regions with diffuse intimal thickening (n=10), fibrotic tissue (n=2) and thick-cap fibroatheroma (n=1) were correctly identified on the classification map. The ability to identify different plaque types using FLIm data alone may serve as a powerful clinical and research tool for studying atherosclerosis in animal models as well as in humans.

Paper Details

Date Published: 27 April 2016
PDF: 1 pages
Proc. SPIE 9689, Photonic Therapeutics and Diagnostics XII, 968934 (27 April 2016); doi: 10.1117/12.2214028
Show Author Affiliations
Hussain Fatakdawala, Univ. of California, Davis (United States)
Dimitris S. Gorpas, Univ. of California, Davis (United States)
Julien Bec, Univ. of California, Davis (United States)
Dinglong M. Ma, Univ. of California, Davis (United States)
Diego R. Yankelevich, Univ. of California, Davis (United States)
John W. Bishop, Univ. of California, Davis (United States)
Laura Marcu, Univ. of California, Davis (United States)

Published in SPIE Proceedings Vol. 9689:
Photonic Therapeutics and Diagnostics XII
Hyun Wook Kang; Guillermo J. Tearney; Melissa C. Skala; Bernard Choi; Andreas Mandelis; Brian J. F. Wong; Justus F. Ilgner; Nikiforos Kollias; Paul J. Campagnola; Kenton W. Gregory; Laura Marcu; Haishan Zeng, Editor(s)

© SPIE. Terms of Use
Back to Top