Biological parametric mapping accounting for random regressors with regression calibration and model II regression

Xue Yang, Carolyn B. Lauzon, Ciprian M Crainiceanu, Brian S Caffo, Susan M. Resnick, Bennett A. Landman

Research output: Contribution to journalArticle

Abstract

Massively univariate regression and inference in the form of statistical parametric mapping have transformed the way in which multi-dimensional imaging data are studied. In functional and structural neuroimaging, the de facto standard "design matrix"-based general linear regression model and its multi-level cousins have enabled investigation of the biological basis of the human brain. With modern study designs, it is possible to acquire multi-modal three-dimensional assessments of the same individuals-e.g., structural, functional and quantitative magnetic resonance imaging, alongside functional and ligand binding maps with positron emission tomography. Largely, current statistical methods in the imaging community assume that the regressors are non-random. For more realistic multi-parametric assessment (e.g., voxel-wise modeling), distributional consideration of all observations is appropriate. Herein, we discuss two unified regression and inference approaches, model II regression and regression calibration, for use in massively univariate inference with imaging data. These methods use the design matrix paradigm and account for both random and non-random imaging regressors. We characterize these methods in simulation and illustrate their use on an empirical dataset. Both methods have been made readily available as a toolbox plug-in for the SPM software.

Original languageEnglish (US)
Pages (from-to)1761-1768
Number of pages8
JournalNeuroImage
Volume62
Issue number3
DOIs
StatePublished - Sep 2012

Fingerprint

Calibration
Linear Models
Functional Neuroimaging
Positron-Emission Tomography
Software
Magnetic Resonance Imaging
Ligands
Brain

Keywords

  • General linear model
  • Model II regression
  • Random regressors
  • Regression calibration
  • Structure-function relationships

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Biological parametric mapping accounting for random regressors with regression calibration and model II regression. / Yang, Xue; Lauzon, Carolyn B.; Crainiceanu, Ciprian M; Caffo, Brian S; Resnick, Susan M.; Landman, Bennett A.

In: NeuroImage, Vol. 62, No. 3, 09.2012, p. 1761-1768.

Research output: Contribution to journalArticle

Yang, Xue ; Lauzon, Carolyn B. ; Crainiceanu, Ciprian M ; Caffo, Brian S ; Resnick, Susan M. ; Landman, Bennett A. / Biological parametric mapping accounting for random regressors with regression calibration and model II regression. In: NeuroImage. 2012 ; Vol. 62, No. 3. pp. 1761-1768.
@article{76f4cf8737544054989ebc364424e06f,
title = "Biological parametric mapping accounting for random regressors with regression calibration and model II regression",
abstract = "Massively univariate regression and inference in the form of statistical parametric mapping have transformed the way in which multi-dimensional imaging data are studied. In functional and structural neuroimaging, the de facto standard {"}design matrix{"}-based general linear regression model and its multi-level cousins have enabled investigation of the biological basis of the human brain. With modern study designs, it is possible to acquire multi-modal three-dimensional assessments of the same individuals-e.g., structural, functional and quantitative magnetic resonance imaging, alongside functional and ligand binding maps with positron emission tomography. Largely, current statistical methods in the imaging community assume that the regressors are non-random. For more realistic multi-parametric assessment (e.g., voxel-wise modeling), distributional consideration of all observations is appropriate. Herein, we discuss two unified regression and inference approaches, model II regression and regression calibration, for use in massively univariate inference with imaging data. These methods use the design matrix paradigm and account for both random and non-random imaging regressors. We characterize these methods in simulation and illustrate their use on an empirical dataset. Both methods have been made readily available as a toolbox plug-in for the SPM software.",
keywords = "General linear model, Model II regression, Random regressors, Regression calibration, Structure-function relationships",
author = "Xue Yang and Lauzon, {Carolyn B.} and Crainiceanu, {Ciprian M} and Caffo, {Brian S} and Resnick, {Susan M.} and Landman, {Bennett A.}",
year = "2012",
month = "9",
doi = "10.1016/j.neuroimage.2012.05.020",
language = "English (US)",
volume = "62",
pages = "1761--1768",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - Biological parametric mapping accounting for random regressors with regression calibration and model II regression

AU - Yang, Xue

AU - Lauzon, Carolyn B.

AU - Crainiceanu, Ciprian M

AU - Caffo, Brian S

AU - Resnick, Susan M.

AU - Landman, Bennett A.

PY - 2012/9

Y1 - 2012/9

N2 - Massively univariate regression and inference in the form of statistical parametric mapping have transformed the way in which multi-dimensional imaging data are studied. In functional and structural neuroimaging, the de facto standard "design matrix"-based general linear regression model and its multi-level cousins have enabled investigation of the biological basis of the human brain. With modern study designs, it is possible to acquire multi-modal three-dimensional assessments of the same individuals-e.g., structural, functional and quantitative magnetic resonance imaging, alongside functional and ligand binding maps with positron emission tomography. Largely, current statistical methods in the imaging community assume that the regressors are non-random. For more realistic multi-parametric assessment (e.g., voxel-wise modeling), distributional consideration of all observations is appropriate. Herein, we discuss two unified regression and inference approaches, model II regression and regression calibration, for use in massively univariate inference with imaging data. These methods use the design matrix paradigm and account for both random and non-random imaging regressors. We characterize these methods in simulation and illustrate their use on an empirical dataset. Both methods have been made readily available as a toolbox plug-in for the SPM software.

AB - Massively univariate regression and inference in the form of statistical parametric mapping have transformed the way in which multi-dimensional imaging data are studied. In functional and structural neuroimaging, the de facto standard "design matrix"-based general linear regression model and its multi-level cousins have enabled investigation of the biological basis of the human brain. With modern study designs, it is possible to acquire multi-modal three-dimensional assessments of the same individuals-e.g., structural, functional and quantitative magnetic resonance imaging, alongside functional and ligand binding maps with positron emission tomography. Largely, current statistical methods in the imaging community assume that the regressors are non-random. For more realistic multi-parametric assessment (e.g., voxel-wise modeling), distributional consideration of all observations is appropriate. Herein, we discuss two unified regression and inference approaches, model II regression and regression calibration, for use in massively univariate inference with imaging data. These methods use the design matrix paradigm and account for both random and non-random imaging regressors. We characterize these methods in simulation and illustrate their use on an empirical dataset. Both methods have been made readily available as a toolbox plug-in for the SPM software.

KW - General linear model

KW - Model II regression

KW - Random regressors

KW - Regression calibration

KW - Structure-function relationships

UR - http://www.scopus.com/inward/record.url?scp=84863521211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863521211&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2012.05.020

DO - 10.1016/j.neuroimage.2012.05.020

M3 - Article

C2 - 22609453

AN - SCOPUS:84863521211

VL - 62

SP - 1761

EP - 1768

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -