Lacunarity-based inherent texture correlation approach for wireless capsule endoscopy image analysis

V. S. Charisis, L. J. Hadjileontiadis, G. D. Sergiadis

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Wireless capsule endoscopy (WCE) is a novel technology that offers non-invasive visual inspection of the digestive tract (DT) and especially small bowel. However, the revision of the large amount of images produced is highly timeconsuming and prone to human error. This weakness was the rationale to propose a novel strategy for automatic detection of WCE images related to ulcer, one of the most common findings of DT. This paper introduces a new texture extraction method based on the inherent texture correlation measure and lacunarity index. WCE data are pre-processed with Bidimensional Ensemble Empirical Mode Decomposition to reveal their congenital structure primitives, and also to adaptively reconstruct new refined images. Classification results demonstrated promising classification accuracy (89.1%) exhibiting high potential towards further research in this field.

Original languageBritish English
Title of host publication13th Mediterranean Conference on Medical and Biological Engineering and Computing 2013 - MEDICON 2013
PublisherSpringer Verlag
Pages297-300
Number of pages4
ISBN (Print)9783319008455
DOIs
StatePublished - 2014
Event13th Mediterranean Conference on Medical and Biological Engineering and Computing 2013, MEDICON 2013 - Seville, Spain
Duration: 25 Sep 201328 Sep 2013

Publication series

NameIFMBE Proceedings
Volume41
ISSN (Print)1680-0737

Conference

Conference13th Mediterranean Conference on Medical and Biological Engineering and Computing 2013, MEDICON 2013
Country/TerritorySpain
CitySeville
Period25/09/1328/09/13

Keywords

  • Feature extraction
  • Multiscale analysis
  • Ulcer

Fingerprint

Dive into the research topics of 'Lacunarity-based inherent texture correlation approach for wireless capsule endoscopy image analysis'. Together they form a unique fingerprint.

Cite this