Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Functional Network Identification in Human Resting-state fMRI using Hierarchical Clustering by Time-Lagged Correlations.
KTH, School of Computer Science and Communication (CSC).
2011 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Mapping of human resting-state functional network through functional magnetic resonance imaging (fMRI) have been prolific in recent years congruent with the maturing of Blood Oxygen Level Dependent (BOLD) imaging. Whereas focus have been maintained upon advancing the complexity of the models with model-free clustering approaches and methods using spectral decomposition, relatively few works have been concentrating on the temporal dimension of fMRI data. Specially lacking is models which include temporal lag of fMRI time series explicitly into account. In this investigation, a framework was developed to map human resting-state functional networks with a model which includes temporal lag. Time-lagged cross correlations and maximum delay times were used as a basis for the construction of distance measures used for hierarchical clustering. The delay maps generated when calculating the cross correlations give preliminary indication that the zero-lag assumption of many current models may be lacking in various aspects. The resulting clusters are spatially noisy, which was expected since the distance measure used do not hold spatial information explicitly. However, multiple structures could be identified from the resulting clusters which strengthens causality between BOLD signal correlations and functional networks. There are significant differences between the clusters from the time-lagged correlation distances and their zero lag counterparts. While the distance maps are significantly different in visualization, some structures may be found in both while some are only identifiable in one. It is concluded that time-lagged correlations could be a useful distance measure in identifying functional networks and structures which could not be identified when assuming zero lag.

Abstract [sv]

Utveckling av funktionell magnetresonans (fMRI, functional Magnetic Resonance Imaging) baserad på mätning av blodets syremättnad i hjärnan (Blood Oxygen Level Dependent, BOLD) har gett forskare nya metoder för kartläggning av hjärnans funktionella nätverk hos människor och djur. BOLD-tekniken mäter blodets syresättning som påverkas av neural aktivitet. Forskning angående metodutvecking har tidigare fokuserat på modeller för att minimera empiriska parametrar från användarsidan och optimera existerande algoritmer för att kunna applicera traditionella analysalgoritmer interaktivt. Relativt få har lagt fokus på den temporala aspekten av fMRI data.

I denna undersökning designerar vi en ny modell som tar hänsyn till tidsfördröjningar som kan finnas i fMRI data. Modellen implementeras som en del av ett komplett analysprogram för hantering av fMRI data. För att hitta funktionella nätverk används en hierarkisk klusteralgorithm med ett distansmått baserat på maximal korrelation med avseende på olika tidsfördröjningar.

Jämnfört med kluster från en distansmatris baserad på nollfördröjningskorrelation, kan vår metod generellt identifiera fler strukturer och funktionella nätverk som är konsistenta med etablerad literatur. Detta förstärker kausaliteten mellan korrelation i fMRI tidserier och funktionella nätverk. Vi drar slutsatsen att en modell som tar i hänsyn tidsfördröjningar kan innebära en förbättring av metodik för analys av fMRI data.

Place, publisher, year, edition, pages
2011.
Series
Trita-CSC-E, ISSN 1653-5715 ; 2011:048
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-130785OAI: oai:DiVA.org:kth-130785DiVA: diva2:654232
Educational program
Master of Science in Engineering -Engineering Physics
Uppsok
Technology
Supervisors
Examiners
Available from: 2013-10-07 Created: 2013-10-07

Open Access in DiVA

No full text

Other links

http://www.nada.kth.se/utbildning/grukth/exjobb/rapportlistor/2011/rapporter11/wang_yanlu_11048.pdf
By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 24 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf