A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions

Gina Maria Pomann, Ana Maria Staicu, Edgar J. Lobaton, Amanda F. Mejia, Blake E. Dewey, Daniel S. Reich, Elizabeth M. Sweeney, Russell T. Shinohara

Research output: Contribution to journalArticle

Abstract

We propose a lag functional linear model to predict a response using multiple functional predictors observed at discrete grids with noise. Two procedures are proposed to estimate the regression parameter functions: (1) an approach that ensures smoothness for each value of time using generalized cross-validation; and (2) a global smoothing approach using a restricted maximum likelihood framework. Numerical studies are presented to analyze predictive accuracy in many realistic scenarios. The methods are employed to estimate a magnetic resonance imaging (MRI)-based measure of tissue damage (the magnetization transfer ratio, or MTR) in multiple sclerosis (MS) lesions, a disease that causes damage to the myelin sheaths around axons in the central nervous system. Our method of estimation of MTR within lesions is useful retrospectively in research applications where MTR was not acquired, as well as in clinical practice settings where acquiring MTR is not currently part of the standard of care. The model facilitates the use of commonly acquired imaging modalities to estimate MTR within lesions, and outperforms cross-sectional models that do not account for temporal patterns of lesion development and repair.

Original languageEnglish (US)
Pages (from-to)2325-2348
Number of pages24
JournalAnnals of Applied Statistics
Volume10
Issue number4
DOIs
StatePublished - Dec 1 2016
Externally publishedYes

Fingerprint

Functional Linear Model
Multiple Sclerosis
Magnetization
Prediction
Damage
Estimate
Multiple Responses
Imaging techniques
Restricted Maximum Likelihood
Generalized Cross-validation
Magnetic Resonance Imaging
Neurology
Magnetic resonance
Modality
Maximum likelihood
Repair
Smoothing
Numerical Study
Predictors
Smoothness

Keywords

  • Functional data analysis
  • Functional linear model
  • Image analysis
  • Magnetization transfer ratio

ASJC Scopus subject areas

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty

Cite this

Pomann, G. M., Staicu, A. M., Lobaton, E. J., Mejia, A. F., Dewey, B. E., Reich, D. S., ... Shinohara, R. T. (2016). A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions. Annals of Applied Statistics, 10(4), 2325-2348. https://doi.org/10.1214/16-AOAS981

A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions. / Pomann, Gina Maria; Staicu, Ana Maria; Lobaton, Edgar J.; Mejia, Amanda F.; Dewey, Blake E.; Reich, Daniel S.; Sweeney, Elizabeth M.; Shinohara, Russell T.

In: Annals of Applied Statistics, Vol. 10, No. 4, 01.12.2016, p. 2325-2348.

Research output: Contribution to journalArticle

Pomann, GM, Staicu, AM, Lobaton, EJ, Mejia, AF, Dewey, BE, Reich, DS, Sweeney, EM & Shinohara, RT 2016, 'A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions', Annals of Applied Statistics, vol. 10, no. 4, pp. 2325-2348. https://doi.org/10.1214/16-AOAS981
Pomann, Gina Maria ; Staicu, Ana Maria ; Lobaton, Edgar J. ; Mejia, Amanda F. ; Dewey, Blake E. ; Reich, Daniel S. ; Sweeney, Elizabeth M. ; Shinohara, Russell T. / A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions. In: Annals of Applied Statistics. 2016 ; Vol. 10, No. 4. pp. 2325-2348.
@article{aa991a5d5a074395badb538656e71716,
title = "A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions",
abstract = "We propose a lag functional linear model to predict a response using multiple functional predictors observed at discrete grids with noise. Two procedures are proposed to estimate the regression parameter functions: (1) an approach that ensures smoothness for each value of time using generalized cross-validation; and (2) a global smoothing approach using a restricted maximum likelihood framework. Numerical studies are presented to analyze predictive accuracy in many realistic scenarios. The methods are employed to estimate a magnetic resonance imaging (MRI)-based measure of tissue damage (the magnetization transfer ratio, or MTR) in multiple sclerosis (MS) lesions, a disease that causes damage to the myelin sheaths around axons in the central nervous system. Our method of estimation of MTR within lesions is useful retrospectively in research applications where MTR was not acquired, as well as in clinical practice settings where acquiring MTR is not currently part of the standard of care. The model facilitates the use of commonly acquired imaging modalities to estimate MTR within lesions, and outperforms cross-sectional models that do not account for temporal patterns of lesion development and repair.",
keywords = "Functional data analysis, Functional linear model, Image analysis, Magnetization transfer ratio",
author = "Pomann, {Gina Maria} and Staicu, {Ana Maria} and Lobaton, {Edgar J.} and Mejia, {Amanda F.} and Dewey, {Blake E.} and Reich, {Daniel S.} and Sweeney, {Elizabeth M.} and Shinohara, {Russell T.}",
year = "2016",
month = "12",
day = "1",
doi = "10.1214/16-AOAS981",
language = "English (US)",
volume = "10",
pages = "2325--2348",
journal = "Annals of Applied Statistics",
issn = "1932-6157",
publisher = "Institute of Mathematical Statistics",
number = "4",

}

TY - JOUR

T1 - A lag functional linear model for prediction of magnetization transfer ratio in multiple sclerosis lesions

AU - Pomann, Gina Maria

AU - Staicu, Ana Maria

AU - Lobaton, Edgar J.

AU - Mejia, Amanda F.

AU - Dewey, Blake E.

AU - Reich, Daniel S.

AU - Sweeney, Elizabeth M.

AU - Shinohara, Russell T.

PY - 2016/12/1

Y1 - 2016/12/1

N2 - We propose a lag functional linear model to predict a response using multiple functional predictors observed at discrete grids with noise. Two procedures are proposed to estimate the regression parameter functions: (1) an approach that ensures smoothness for each value of time using generalized cross-validation; and (2) a global smoothing approach using a restricted maximum likelihood framework. Numerical studies are presented to analyze predictive accuracy in many realistic scenarios. The methods are employed to estimate a magnetic resonance imaging (MRI)-based measure of tissue damage (the magnetization transfer ratio, or MTR) in multiple sclerosis (MS) lesions, a disease that causes damage to the myelin sheaths around axons in the central nervous system. Our method of estimation of MTR within lesions is useful retrospectively in research applications where MTR was not acquired, as well as in clinical practice settings where acquiring MTR is not currently part of the standard of care. The model facilitates the use of commonly acquired imaging modalities to estimate MTR within lesions, and outperforms cross-sectional models that do not account for temporal patterns of lesion development and repair.

AB - We propose a lag functional linear model to predict a response using multiple functional predictors observed at discrete grids with noise. Two procedures are proposed to estimate the regression parameter functions: (1) an approach that ensures smoothness for each value of time using generalized cross-validation; and (2) a global smoothing approach using a restricted maximum likelihood framework. Numerical studies are presented to analyze predictive accuracy in many realistic scenarios. The methods are employed to estimate a magnetic resonance imaging (MRI)-based measure of tissue damage (the magnetization transfer ratio, or MTR) in multiple sclerosis (MS) lesions, a disease that causes damage to the myelin sheaths around axons in the central nervous system. Our method of estimation of MTR within lesions is useful retrospectively in research applications where MTR was not acquired, as well as in clinical practice settings where acquiring MTR is not currently part of the standard of care. The model facilitates the use of commonly acquired imaging modalities to estimate MTR within lesions, and outperforms cross-sectional models that do not account for temporal patterns of lesion development and repair.

KW - Functional data analysis

KW - Functional linear model

KW - Image analysis

KW - Magnetization transfer ratio

UR - http://www.scopus.com/inward/record.url?scp=85009231253&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009231253&partnerID=8YFLogxK

U2 - 10.1214/16-AOAS981

DO - 10.1214/16-AOAS981

M3 - Article

AN - SCOPUS:85009231253

VL - 10

SP - 2325

EP - 2348

JO - Annals of Applied Statistics

JF - Annals of Applied Statistics

SN - 1932-6157

IS - 4

ER -