Extended multi-resolution local patterns - A discriminative feature learning approach for colonoscopy image classification

Siyamalan Manivannan (Lead / Corresponding author), Manuel Trucco

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a novel local image descriptor called the Extended Multi-resolution Local Patterns, and a discriminative probabilistic framework for learning its parameters together with a multi-class image classifier. Our approach uses training data with image-level labels to learn the features which are discriminative for multi-class colonoscopy image classification. Experiments on a three class (abnormal, normal, uninformative) white-light colonoscopy image dataset with 2800 images show that the proposed feature perform better than popular hand- designed features used in the medical as well as in the computer vision literature for image classification.

Original languageEnglish
Title of host publicationCARE 2016
Subtitle of host publicationComputer-Assisted and Robotic Endoscopy
EditorsGuang-Zhong Yang, Nassir Navab, Jonathan McLeod, Terry Peters, Kensaku Mori, Xiongbiao Luo, Tobias Reichl
Place of PublicationSwitzerland
PublisherSpringer
Pages48-58
Number of pages11
Volume10170
ISBN (Electronic)9783319540573
ISBN (Print)9783319540566
DOIs
Publication statusPublished - 2017
Event2nd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 - Athens, Greece
Duration: 17 Oct 201617 Oct 2016

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume10170
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016
CountryGreece
CityAthens
Period17/10/1617/10/16

    Fingerprint

Keywords

  • Local Binary Pattern
  • Adenoma Detection Rate
  • Local Ternary Pattern
  • Soft Label
  • Colon Dataset

Cite this

Manivannan, S., & Trucco, M. (2017). Extended multi-resolution local patterns - A discriminative feature learning approach for colonoscopy image classification. In G-Z. Yang, N. Navab, J. McLeod, T. Peters, K. Mori, X. Luo, & T. Reichl (Eds.), CARE 2016: Computer-Assisted and Robotic Endoscopy (Vol. 10170, pp. 48-58). (Lecture Notes in Computer Science; Vol. 10170). Springer . https://doi.org/10.1007/978-3-319-54057-3_5