Boosting connectome classification via combination of geometric and topological normalizations

Dmitry Petrov, Yulia Dodonova, Leonid Zhukov, Mikhail Belyaev

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

4 Citations (Scopus)

Abstract

The structural connectome classification is a challenging task due to a small sample size and high dimensionality of feature space. In this paper, we propose a new data prepossessing method that combines geometric and topological connectome normalization and significantly improves classification results. We validate this approach by performing classification between autism spectrum disorder and normal development connectomes in children and adolescents. We demonstrate a significant enhancement in performance using weighted and normalized data over the best available model (boosted decision trees) trained on baseline features.

Original languageEnglish
Title of host publicationPRNI 2016 - 6th International Workshop on Pattern Recognition in Neuroimaging
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467365307
DOIs
Publication statusPublished - 24 Aug 2016
Externally publishedYes
Event6th International Workshop on Pattern Recognition in Neuroimaging, PRNI 2016 - Trento, Italy
Duration: 22 Jun 201624 Jun 2016

Publication series

NamePRNI 2016 - 6th International Workshop on Pattern Recognition in Neuroimaging

Conference

Conference6th International Workshop on Pattern Recognition in Neuroimaging, PRNI 2016
Country/TerritoryItaly
CityTrento
Period22/06/1624/06/16

Keywords

  • Autism spectrum disorder
  • Brain networks
  • Diffusion tensor imaging
  • Graph theory
  • Machine Learning

Fingerprint

Dive into the research topics of 'Boosting connectome classification via combination of geometric and topological normalizations'. Together they form a unique fingerprint.

Cite this