Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network

Sukesh Kumar Das, Pratik Jain, Anil K. Sao, Bharat Biswal

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


Blood Oxygen Level-Dependent (BOLD) time course in functional magnetic resonance imaging (fMRI) is modeled as the response of the hemodynamic response function (HRF) excited by an activity-inducing signal. Variability of the HRF across the brain influences functional connectivity (FC) estimates and some approaches have been attempted to separate the HRF and activity-inducing signal from the observed BOLD signal as a blind separation problem. In this work, an approach based on homomorphic filtering is proposed to estimate a non-parametric representation of HRF in resting state fMRI. Voxel-wise and region-wise variations of correlation of the estimated HRF (both the parametric and non-parametric representation) are analyzed in different functional networks. Principal component analysis of the correlation matrix using the estimated HRF is used to analyze the interconnectedness. HRF shows higher variability for the non-parametric representation over the parametric representation. Further, the contribution of the estimated HRF is then studied in producing resting-state networks using the dictionary learning framework.

Original languageEnglish (US)
Title of host publicationBrain Informatics - 16th International Conference, BI 2023, Proceedings
EditorsFeng Liu, Hongjun Wang, Yu Zhang, Hongzhi Kuai, Emily P. Stephen
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages10
ISBN (Print)9783031430749
StatePublished - 2023
Event16th International Conference on Brain Informatics, BI 2023 - Hoboken, United States
Duration: Aug 1 2023Aug 3 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13974 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference16th International Conference on Brain Informatics, BI 2023
Country/TerritoryUnited States

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


  • Deconvolution
  • Dictionary learning
  • HRF variability
  • Resting state fMRI
  • Resting state networks


Dive into the research topics of 'Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network'. Together they form a unique fingerprint.

Cite this