Video-specific SVMs for colonoscopy image classification

Siyamalan Manivannan (Lead / Corresponding author), Ruixuan Wang, Maria P. Trujillo, Jesus Arbey Hoyos, Emanuele Trucco

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

2 Citations (Scopus)


We propose a novel classification framework called the videospecific SVM (V-SVM) for normal-vs-abnormal white-light colonoscopy image classification. V-SVM is an ensemble of linear SVMs, with each trained to separate the abnormal images in a particular video from all the normal images in all the videos. Since V-SVM is designed to capture lesion-specific properties as well as intra-class variations it is expected to perform better than SVM. Experiments on a colonoscopy image dataset with about 10,000 images show that V-SVM significantly improves the performance over SVM and other baseline classifiers.

Original languageEnglish
Title of host publicationComputer-Assisted and Robotic Endoscopy
Subtitle of host publicationFirst International Workshop, CARE 2014, held in conjunction with MICCAI 2014, Boston, MA, USA, September 18, 2014. Revised selected papers
EditorsXiongbiao Luo, Tobias Reichl , Daniel Mirota, Timothy Soper
PublisherSpringer International Publishing
Number of pages11
ISBN (Electronic)9783319134109
ISBN (Print)9783319134093
Publication statusPublished - 23 Nov 2014
Event1st International Workshop on Computer-Assisted and Robotic Endoscopy - MIT Campus and at Harvard Medical School, Boston, United States
Duration: 18 Sept 2014 → …

Publication series

NameLecture notes in computer science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Workshop1st International Workshop on Computer-Assisted and Robotic Endoscopy
Abbreviated titleCARE 2014
Country/TerritoryUnited States
Period18/09/14 → …
OtherHeld in Conjunction with MICCAI 2014
Internet address

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science


Dive into the research topics of 'Video-specific SVMs for colonoscopy image classification'. Together they form a unique fingerprint.

Cite this