DMCA
Endoscopic Video Manifolds for Targeted Optical Biopsy (2012)
| Venue: | IEEE Trans. on Medical Imaging |
| Citations: | 2 - 0 self |
BibTeX
@ARTICLE{Atasoy12endoscopicvideo,
author = {Selen Atasoy and Diana Mateus and Er Meining and Guang-zhong Yang and Nassir Navab},
title = {Endoscopic Video Manifolds for Targeted Optical Biopsy},
journal = {IEEE Trans. on Medical Imaging},
year = {2012},
pages = {637--653}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract—Gastro-intestinal (GI) endoscopy is a widely used clinical procedure for screening and surveillance of digestive tract diseases ranging from Barrett’s Oesophagus to oesophageal cancer. Current surveillance protocol consists of periodic endo-scopic examinations performed in 3–4 month intervals including expert’s visual assessment and biopsies taken from suspicious tissue regions. Recent development of a new imaging technology, called probe-based confocal laser endomicroscopy (pCLE), en-abled the acquisition of in vivo optical biopsies without removing any tissue sample. Besides its several advantages, i.e., noninva-siveness, real-time and in vivo feedback, optical biopsies involve a new challenge for the endoscopic expert. Due to their noninva-sive nature, optical biopsies do not leave any scar on the tissue and therefore recognition of the previous optical biopsy sites in surveillance endoscopy becomes very challenging. In this work, we introduce a clustering and classification framework to facilitate retargeting previous optical biopsy sites in surveillance upper GI-endoscopies. A new representation of endoscopic videos based on manifold learning, “endoscopic video manifolds ” (EVMs), is proposed. The low dimensional EVM representation is adapted to facilitate two different clustering tasks; i.e., clustering of in-formative frames and patient specific endoscopic segments, only by changing the similarity measure. Each step of the proposed framework is validated on three in vivo patient datasets containing 1834, 3445, and 1546 frames, corresponding to endoscopic videos of 73.36, 137.80, and 61.84 s, respectively. Improvements achieved by the introduced EVM representation are demonstrated by quantitative analysis in comparison to the original image repre-sentation and principal component analysis. Final experiments evaluating the complete framework demonstrate the feasibility of the proposed method as a promising step for assisting the endoscopic expert in retargeting the optical biopsy sites. Index Terms—Classification, clustering, gastro-intestinal (GI)-endoscopy, manifold learning, optical biopsy.
Keyphrases
targeted optical biopsy optical biopsy endoscopic video manifold manifold learning endoscopic expert previous optical biopsy site suspicious tissue region periodic endo-scopic examination new imaging technology endoscopic video final experiment vivo optical biopsy principal component analysis low dimensional evm representation digestive tract disease therefore recognition surveillance endoscopy barrett oesophagus classification framework video manifold in-formative frame patient specific endoscopic segment current surveillance protocol tissue sample promising step surveillance upper gi-endoscopies expert visual assessment noninva-sive nature probe-based confocal laser endomicroscopy quantitative analysis new representation recent development optical biopsy site index term classification introduced evm representation abstract gastro-intestinal vivo feedback vivo patient datasets clinical procedure new challenge similarity measure complete framework original image repre-sentation month interval several advantage
