Self-reproduction versus transition rules in ant colonies for medical volume segmentation

Research output: Contribution to book/Conference proceedings/Anthology/Report › Conference contribution › Contributed › peer-review

Contributors

  • Robert Haase - , University Medicine (Faculty of Medicine and University Hospital) (Author)
  • Hans Joachim Böhme - , Dresden University of Applied Sciences (HTW) (Author)
  • Rosalind Perrin - , Medical Faculty Carl Gustav Carus (Author)
  • Klaus Zöphel - , TUD Dresden University of Technology (Author)
  • Nasreddin Abolmaali - , Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology (Author)

Abstract

Target volume delineation in image stacks resulting from low contrast positron emission tomography (PET) remains a hot topic in the field of medical image processing. We propose an algorithm based on artificial ants moving in three dimensional image space controlled by transition rules which are able to self-reproduce. This investigation shows by variation of the transition rules that the impact on segmentation results is small because self-reproduction is the overwhelming effect in the simulation.

Details

Original languageEnglish
Title of host publicationSwarm Intelligence
PublisherSpringer, Berlin [u. a.]
Pages316-323
Number of pages8
ISBN (print)978-3-642-32649-3
Publication statusPublished - 2012
Peer-reviewedYes

Publication series

SeriesLecture Notes in Computer Science, Volume 7461
ISSN0302-9743

Conference

Title8th International Conference on Swarm Intelligence
Abbreviated titleANTS 2012
Conference number8
Duration12 - 14 September 2012
LocationUniversite Libre de Bruxelles
CityBrussels
CountryBelgium

External IDs

WOS 000339488700033

Keywords

ASJC Scopus subject areas

Keywords

  • Algorithm

Library keywords