Colour image analysis of wireless capsule endoscopy video: A review

Fisher, Mark and Mackiewicz, Michal ORCID: (2013) Colour image analysis of wireless capsule endoscopy video: A review. In: Color Medical Image Analysis, Lecture Notes in Computational Vision and Biomechanics. Springer. ISBN 978-94-017-8129-9

Full text not available from this repository. (Request a copy)


Wireless capsule endoscopy (CE) has been available since 2001 and is now established as one of the principal approaches used to examine the small bowel, with a range of devices available from four manufacturers. But although its use is widespread the reading of CE videos remains an arduous and time consuming exercise for gastroenterologists because relevant frames which are of interest to the physician constitute only about 1\% of the video. CE exam viewing times vary from 40-90 minutes, depending on the clinician's experience, the complexity of the case and the small bowel transit time. Colour image analysis has been applied by manufacturers to speed up this process, for example, Given Imaging's \emph{Rapid Reader} Software includes a suspected blood indicator (SBI) designed to detect bleeding in the video. However, some evaluations of this tool reported concerns with regard to its specificity and sensitivity and so it does not free the specialist from reviewing the entire footage and can only be used as a fast screening aid. Over the past decade a number of papers have proposed a range of colour image processing and computer vision applications to assist the gastroenterologist in the analysis of CE video. These techniques can be divided into three categories, the first considers the topographic segmentation of CE video into meaningful parts such as mouth, oesophagus, stomach, small intestine, and colon. The second involves the detection of clinically significant video events (both abnormal and normal) and the third attempts to adaptively adjust the video viewing speed. This chapter reviews this research focusing particularly on the role of colour and texture descriptors and concludes by suggesting possible future directions for CE analysis.

Item Type: Book Section
Faculty \ School: Faculty of Science > School of Computing Sciences
UEA Research Groups: Faculty of Science > Research Groups > Interactive Graphics and Audio
Faculty of Science > Research Groups > Colour and Imaging Lab
Depositing User: Mark Fisher
Date Deposited: 07 May 2013 13:44
Last Modified: 17 Sep 2023 06:31
DOI: 10.1007/978-94-007-5389-1_7

Actions (login required)

View Item View Item