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

doi:10.1007/978-3-031-43075-6_21

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 proceeding › Conference contribution

Abstract

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 language	English (US)
Title of host publication	Brain Informatics - 16th International Conference, BI 2023, Proceedings
Editors	Feng Liu, Hongjun Wang, Yu Zhang, Hongzhi Kuai, Emily P. Stephen
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	239-248
Number of pages	10
ISBN (Print)	9783031430749
DOIs	https://doi.org/10.1007/978-3-031-43075-6_21
State	Published - 2023
Event	16th International Conference on Brain Informatics, BI 2023 - Hoboken, United States Duration: Aug 1 2023 → Aug 3 2023

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13974 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	16th International Conference on Brain Informatics, BI 2023
Country/Territory	United States
City	Hoboken
Period	8/1/23 → 8/3/23

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Keywords

Deconvolution
Dictionary learning
HRF variability
Resting state fMRI
Resting state networks

Access to Document

10.1007/978-3-031-43075-6_21

Cite this

Das, S. K., Jain, P., Sao, A. K., & Biswal, B. (2023). Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network. In F. Liu, H. Wang, Y. Zhang, H. Kuai, & E. P. Stephen (Eds.), Brain Informatics - 16th International Conference, BI 2023, Proceedings (pp. 239-248). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13974 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-43075-6_21

Das, Sukesh Kumar ; Jain, Pratik ; Sao, Anil K. et al. / Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network. Brain Informatics - 16th International Conference, BI 2023, Proceedings. editor / Feng Liu ; Hongjun Wang ; Yu Zhang ; Hongzhi Kuai ; Emily P. Stephen. Springer Science and Business Media Deutschland GmbH, 2023. pp. 239-248 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{75563ba2e353480892e58447883df4fc,

title = "Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network",

abstract = "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.",

keywords = "Deconvolution, Dictionary learning, HRF variability, Resting state fMRI, Resting state networks",

author = "Das, {Sukesh Kumar} and Pratik Jain and Sao, {Anil K.} and Bharat Biswal",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 16th International Conference on Brain Informatics, BI 2023 ; Conference date: 01-08-2023 Through 03-08-2023",

year = "2023",

doi = "10.1007/978-3-031-43075-6_21",

language = "English (US)",

isbn = "9783031430749",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "239--248",

editor = "Feng Liu and Hongjun Wang and Yu Zhang and Hongzhi Kuai and Stephen, {Emily P.}",

booktitle = "Brain Informatics - 16th International Conference, BI 2023, Proceedings",

address = "Germany",

}

Das, SK, Jain, P, Sao, AK & Biswal, B 2023, Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network. in F Liu, H Wang, Y Zhang, H Kuai & EP Stephen (eds), Brain Informatics - 16th International Conference, BI 2023, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13974 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 239-248, 16th International Conference on Brain Informatics, BI 2023, Hoboken, United States, 8/1/23. https://doi.org/10.1007/978-3-031-43075-6_21

Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network. / Das, Sukesh Kumar; Jain, Pratik; Sao, Anil K. et al.
Brain Informatics - 16th International Conference, BI 2023, Proceedings. ed. / Feng Liu; Hongjun Wang; Yu Zhang; Hongzhi Kuai; Emily P. Stephen. Springer Science and Business Media Deutschland GmbH, 2023. p. 239-248 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13974 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network

AU - Das, Sukesh Kumar

AU - Jain, Pratik

AU - Sao, Anil K.

AU - Biswal, Bharat

PY - 2023

Y1 - 2023

N2 - 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.

AB - 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.

KW - Deconvolution

KW - Dictionary learning

KW - HRF variability

KW - Resting state fMRI

KW - Resting state networks

UR - http://www.scopus.com/inward/record.url?scp=85172416379&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85172416379&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-43075-6_21

DO - 10.1007/978-3-031-43075-6_21

M3 - Conference contribution

AN - SCOPUS:85172416379

SN - 9783031430749

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 239

EP - 248

BT - Brain Informatics - 16th International Conference, BI 2023, Proceedings

A2 - Liu, Feng

A2 - Wang, Hongjun

A2 - Zhang, Yu

A2 - Kuai, Hongzhi

A2 - Stephen, Emily P.

PB - Springer Science and Business Media Deutschland GmbH

T2 - 16th International Conference on Brain Informatics, BI 2023

Y2 - 1 August 2023 through 3 August 2023

ER -

Das SK, Jain P, Sao AK, Biswal B. Variability of Non-parametric HRF in Interconnectedness and Its Association in Deriving Resting State Network. In Liu F, Wang H, Zhang Y, Kuai H, Stephen EP, editors, Brain Informatics - 16th International Conference, BI 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 239-248. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-43075-6_21