Selection of functionally homogeneous brain regions based on correlation-clustering analysis

Stanislav Kozlov, Alexey Poyda, Vyacheslav Orlov, Denis Malakhov, Vadim Ushakov, Maxim Sharaev

    Research output: Contribution to journalConference articlepeer-review

    5 Citations (Scopus)

    Abstract

    In region of interest (ROI) brain analysis, the proper selection of voxels in ROIs plays a crucial role. In existing methods for selection of functionally homogeneous regions of human brain based on fMRI data, each voxel is attributed to some brain region, not taking into account the possibility of the existence of borderline voxels demonstrating transitional dynamics that cannot be clearly attributed to any of the regions between which these voxels are located. As a result, the situation when voxels formally assigned to one region, but located on the opposite borders of the region, have a correlation close to zero or even negative. In such cases the aggregation of voxels activity does not reflect properly the behavior of the whole area. In this article we present two methods for identifying functionally homogeneous regions based on fMRI data, using a correlation-cluster approach: One of the methods allows identifying functionally homogeneous regions, where voxels have a high level of correlation, while the second one allows identifying functionally homogeneous regions, which are stable over time. Both methods assume that not all brain voxels will be assigned to regions.

    Original languageEnglish
    Pages (from-to)519-526
    Number of pages8
    JournalProcedia Computer Science
    Volume169
    DOIs
    Publication statusPublished - 2020
    Event10th Annual International Conference on Biologically Inspired Cognitive Architectures, BICA 2019 - Seattle, United States
    Duration: 15 Aug 201919 Aug 2019

    Keywords

    • fMRI
    • functionally homogeneous areas
    • neuroimaging
    • ROI analysis

    Fingerprint

    Dive into the research topics of 'Selection of functionally homogeneous brain regions based on correlation-clustering analysis'. Together they form a unique fingerprint.

    Cite this