A nonparametric Bayesian model for estimating spectral densities of resting-state EEG twin data

Brian Hart; Michele Guindani; Stephen Malone; Mark Fiecas

doi:10.1111/biom.13393

A nonparametric Bayesian model for estimating spectral densities of resting-state EEG twin data

Brian Hart, Michele Guindani, Stephen Malone, Mark Fiecas

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

Electroencephalography (EEG) is a noninvasive neuroimaging modality that captures electrical brain activity many times per second. We seek to estimate power spectra from EEG data that ware gathered for 557 adolescent twin pairs through the Minnesota Twin Family Study (MTFS). Typically, spectral analysis methods treat time series from each subject separately, and independent spectral densities are fit to each time series. Since the EEG data were collected on twins, it is reasonable to assume that the time series have similar underlying characteristics, so borrowing information across subjects can significantly improve estimation. We propose a Nested Bernstein Dirichlet prior model to estimate the power spectrum of the EEG signal for each subject by smoothing periodograms within and across subjects while requiring minimal user input to tuning parameters. Furthermore, we leverage the MTFS twin study design to estimate the heritability of EEG power spectra with the hopes of establishing new endophenotypes. Through simulation studies designed to mimic the MTFS, we show our method out-performs a set of other popular methods.

Original language	English (US)
Pages (from-to)	313-323
Number of pages	11
Journal	Biometrics
Volume	78
Issue number	1
DOIs	https://doi.org/10.1111/biom.13393
State	Published - Mar 2022

Bibliographical note

Funding Information:
Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919-01A, NIH R37 DA05147 and NIH R37 AA09367. Work on this paper by Michele Guindani was supported by NSF SES-1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity.

Funding Information:
Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919‐01A. Work on this paper by Michele Guindani was supported by NSF SES‐1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity.

Publisher Copyright:
© 2020 The International Biometric Society.

Keywords

Bernstein polynomial
Whittle likelihood
heritability
nested Dirichlet process
time series

PubMed: MeSH publication types

Journal Article
Twin Study
Research Support, U.S. Gov't, Non-P.H.S.
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

Access

10.1111/biom.13393

OpenUrl availability

Full text

Cite this

@article{1acfd571ea7e41fe990c9ff43271101b,

title = "A nonparametric Bayesian model for estimating spectral densities of resting-state EEG twin data",

abstract = "Electroencephalography (EEG) is a noninvasive neuroimaging modality that captures electrical brain activity many times per second. We seek to estimate power spectra from EEG data that ware gathered for 557 adolescent twin pairs through the Minnesota Twin Family Study (MTFS). Typically, spectral analysis methods treat time series from each subject separately, and independent spectral densities are fit to each time series. Since the EEG data were collected on twins, it is reasonable to assume that the time series have similar underlying characteristics, so borrowing information across subjects can significantly improve estimation. We propose a Nested Bernstein Dirichlet prior model to estimate the power spectrum of the EEG signal for each subject by smoothing periodograms within and across subjects while requiring minimal user input to tuning parameters. Furthermore, we leverage the MTFS twin study design to estimate the heritability of EEG power spectra with the hopes of establishing new endophenotypes. Through simulation studies designed to mimic the MTFS, we show our method out-performs a set of other popular methods.",

keywords = "Bernstein polynomial, Whittle likelihood, heritability, nested Dirichlet process, time series",

author = "Brian Hart and Michele Guindani and Stephen Malone and Mark Fiecas",

note = "Funding Information: Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919-01A, NIH R37 DA05147 and NIH R37 AA09367. Work on this paper by Michele Guindani was supported by NSF SES-1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity. Funding Information: Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919‐01A. Work on this paper by Michele Guindani was supported by NSF SES‐1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity. Publisher Copyright: {\textcopyright} 2020 The International Biometric Society.",

year = "2022",

month = mar,

doi = "10.1111/biom.13393",

language = "English (US)",

volume = "78",

pages = "313--323",

journal = "Biometrics",

issn = "0006-341X",

publisher = "Wiley-Blackwell",

number = "1",

}

TY - JOUR

T1 - A nonparametric Bayesian model for estimating spectral densities of resting-state EEG twin data

AU - Hart, Brian

AU - Guindani, Michele

AU - Malone, Stephen

AU - Fiecas, Mark

N1 - Funding Information: Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919-01A, NIH R37 DA05147 and NIH R37 AA09367. Work on this paper by Michele Guindani was supported by NSF SES-1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity. Funding Information: Computational resources for this work were provided by the Minnesota Supercomputing Institute at the University of Minnesota. This work was funded in part by the University of Minnesota Informatics Institute. Work on this paper by Stephen Malone and Mark Fiecas was supported by NIH grant R21AA026919‐01A. Work on this paper by Michele Guindani was supported by NSF SES‐1659921 grant: Collaborative Research: Bayesian Approaches for Inference on Brain Connectivity. Publisher Copyright: © 2020 The International Biometric Society.

PY - 2022/3

Y1 - 2022/3

N2 - Electroencephalography (EEG) is a noninvasive neuroimaging modality that captures electrical brain activity many times per second. We seek to estimate power spectra from EEG data that ware gathered for 557 adolescent twin pairs through the Minnesota Twin Family Study (MTFS). Typically, spectral analysis methods treat time series from each subject separately, and independent spectral densities are fit to each time series. Since the EEG data were collected on twins, it is reasonable to assume that the time series have similar underlying characteristics, so borrowing information across subjects can significantly improve estimation. We propose a Nested Bernstein Dirichlet prior model to estimate the power spectrum of the EEG signal for each subject by smoothing periodograms within and across subjects while requiring minimal user input to tuning parameters. Furthermore, we leverage the MTFS twin study design to estimate the heritability of EEG power spectra with the hopes of establishing new endophenotypes. Through simulation studies designed to mimic the MTFS, we show our method out-performs a set of other popular methods.

AB - Electroencephalography (EEG) is a noninvasive neuroimaging modality that captures electrical brain activity many times per second. We seek to estimate power spectra from EEG data that ware gathered for 557 adolescent twin pairs through the Minnesota Twin Family Study (MTFS). Typically, spectral analysis methods treat time series from each subject separately, and independent spectral densities are fit to each time series. Since the EEG data were collected on twins, it is reasonable to assume that the time series have similar underlying characteristics, so borrowing information across subjects can significantly improve estimation. We propose a Nested Bernstein Dirichlet prior model to estimate the power spectrum of the EEG signal for each subject by smoothing periodograms within and across subjects while requiring minimal user input to tuning parameters. Furthermore, we leverage the MTFS twin study design to estimate the heritability of EEG power spectra with the hopes of establishing new endophenotypes. Through simulation studies designed to mimic the MTFS, we show our method out-performs a set of other popular methods.

KW - Bernstein polynomial

KW - Whittle likelihood

KW - heritability

KW - nested Dirichlet process

KW - time series

UR - http://www.scopus.com/inward/record.url?scp=85093959466&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85093959466&partnerID=8YFLogxK

U2 - 10.1111/biom.13393

DO - 10.1111/biom.13393

M3 - Article

C2 - 33058149

AN - SCOPUS:85093959466

SN - 0006-341X

VL - 78

SP - 313

EP - 323

JO - Biometrics

JF - Biometrics

IS - 1

ER -

A nonparametric Bayesian model for estimating spectral densities of resting-state EEG twin data

Abstract

Bibliographical note

Keywords

PubMed: MeSH publication types

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this