Analyzing neuroimaging data with subclasses

A shrinkage approach

Johannes Höhne, Daniel Bartz, Martin N. Hebart, Klaus Muller, Benjamin Blankertz

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Among the numerous methods used to analyze neuroimaging data, Linear Discriminant Analysis (LDA) is commonly applied for binary classification problems. LDAs popularity derives from its simplicity and its competitive classification performance, which has been reported for various types of neuroimaging data.Yet the standard LDA approach proves less than optimal for binary classification problems when additional label information (i.e. subclass labels) is present. Subclass labels allow to model structure in the data, which can be used to facilitate the classification task. In this paper, we illustrate how neuroimaging data exhibit subclass labels that may contain valuable information. We also show that the standard LDA classifier is unable to exploit subclass labels. We introduce a novel method that allows subclass labels to be incorporated efficiently into the classifier. The novel method, which we call Relevance Subclass LDA (RSLDA), computes an individual classification hyperplane for each subclass. It is based on regularized estimators of the subclass mean and uses other subclasses as regularization targets. We demonstrate the applicability and performance of our method on data drawn from two different neuroimaging modalities: (I) EEG data from brain-computer interfacing with event-related potentials, and (II) fMRI data in response to different levels of visual motion. We show that RSLDA outperforms the standard LDA approach for both types of datasets. These findings illustrate the benefits of exploiting subclass structure in neuroimaging data. Finally, we show that our classifier also outputs regularization profiles, enabling researchers to interpret the subclass structure in a meaningful way.RSLDA therefore yields increased classification accuracy as well as a better interpretation of neuroimaging data. Since both results are highly favorable, we suggest to apply RSLDA for various classification problems within neuroimaging and beyond.

Original languageEnglish
Pages (from-to)740-751
Number of pages12
JournalNeuroImage
Volume124
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Neuroimaging
Discriminant Analysis
Evoked Potentials
Electroencephalography
Research Personnel
Magnetic Resonance Imaging
Brain

Keywords

  • BCI
  • Data-driven clustering
  • EEG
  • ERP
  • FMRI
  • Linear classifier
  • Pattern classification
  • Regularization profile
  • Searchlight
  • Shrinkage
  • Single-trial classification
  • Subclass

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Höhne, J., Bartz, D., Hebart, M. N., Muller, K., & Blankertz, B. (2016). Analyzing neuroimaging data with subclasses: A shrinkage approach. NeuroImage, 124, 740-751. https://doi.org/10.1016/j.neuroimage.2015.09.031

Analyzing neuroimaging data with subclasses : A shrinkage approach. / Höhne, Johannes; Bartz, Daniel; Hebart, Martin N.; Muller, Klaus; Blankertz, Benjamin.

In: NeuroImage, Vol. 124, 01.01.2016, p. 740-751.

Research output: Contribution to journalArticle

Höhne, J, Bartz, D, Hebart, MN, Muller, K & Blankertz, B 2016, 'Analyzing neuroimaging data with subclasses: A shrinkage approach', NeuroImage, vol. 124, pp. 740-751. https://doi.org/10.1016/j.neuroimage.2015.09.031
Höhne, Johannes ; Bartz, Daniel ; Hebart, Martin N. ; Muller, Klaus ; Blankertz, Benjamin. / Analyzing neuroimaging data with subclasses : A shrinkage approach. In: NeuroImage. 2016 ; Vol. 124. pp. 740-751.
@article{cbef08ca4603419f951aa4abcb1f1d82,
title = "Analyzing neuroimaging data with subclasses: A shrinkage approach",
abstract = "Among the numerous methods used to analyze neuroimaging data, Linear Discriminant Analysis (LDA) is commonly applied for binary classification problems. LDAs popularity derives from its simplicity and its competitive classification performance, which has been reported for various types of neuroimaging data.Yet the standard LDA approach proves less than optimal for binary classification problems when additional label information (i.e. subclass labels) is present. Subclass labels allow to model structure in the data, which can be used to facilitate the classification task. In this paper, we illustrate how neuroimaging data exhibit subclass labels that may contain valuable information. We also show that the standard LDA classifier is unable to exploit subclass labels. We introduce a novel method that allows subclass labels to be incorporated efficiently into the classifier. The novel method, which we call Relevance Subclass LDA (RSLDA), computes an individual classification hyperplane for each subclass. It is based on regularized estimators of the subclass mean and uses other subclasses as regularization targets. We demonstrate the applicability and performance of our method on data drawn from two different neuroimaging modalities: (I) EEG data from brain-computer interfacing with event-related potentials, and (II) fMRI data in response to different levels of visual motion. We show that RSLDA outperforms the standard LDA approach for both types of datasets. These findings illustrate the benefits of exploiting subclass structure in neuroimaging data. Finally, we show that our classifier also outputs regularization profiles, enabling researchers to interpret the subclass structure in a meaningful way.RSLDA therefore yields increased classification accuracy as well as a better interpretation of neuroimaging data. Since both results are highly favorable, we suggest to apply RSLDA for various classification problems within neuroimaging and beyond.",
keywords = "BCI, Data-driven clustering, EEG, ERP, FMRI, Linear classifier, Pattern classification, Regularization profile, Searchlight, Shrinkage, Single-trial classification, Subclass",
author = "Johannes H{\"o}hne and Daniel Bartz and Hebart, {Martin N.} and Klaus Muller and Benjamin Blankertz",
year = "2016",
month = "1",
day = "1",
doi = "10.1016/j.neuroimage.2015.09.031",
language = "English",
volume = "124",
pages = "740--751",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Analyzing neuroimaging data with subclasses

T2 - A shrinkage approach

AU - Höhne, Johannes

AU - Bartz, Daniel

AU - Hebart, Martin N.

AU - Muller, Klaus

AU - Blankertz, Benjamin

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Among the numerous methods used to analyze neuroimaging data, Linear Discriminant Analysis (LDA) is commonly applied for binary classification problems. LDAs popularity derives from its simplicity and its competitive classification performance, which has been reported for various types of neuroimaging data.Yet the standard LDA approach proves less than optimal for binary classification problems when additional label information (i.e. subclass labels) is present. Subclass labels allow to model structure in the data, which can be used to facilitate the classification task. In this paper, we illustrate how neuroimaging data exhibit subclass labels that may contain valuable information. We also show that the standard LDA classifier is unable to exploit subclass labels. We introduce a novel method that allows subclass labels to be incorporated efficiently into the classifier. The novel method, which we call Relevance Subclass LDA (RSLDA), computes an individual classification hyperplane for each subclass. It is based on regularized estimators of the subclass mean and uses other subclasses as regularization targets. We demonstrate the applicability and performance of our method on data drawn from two different neuroimaging modalities: (I) EEG data from brain-computer interfacing with event-related potentials, and (II) fMRI data in response to different levels of visual motion. We show that RSLDA outperforms the standard LDA approach for both types of datasets. These findings illustrate the benefits of exploiting subclass structure in neuroimaging data. Finally, we show that our classifier also outputs regularization profiles, enabling researchers to interpret the subclass structure in a meaningful way.RSLDA therefore yields increased classification accuracy as well as a better interpretation of neuroimaging data. Since both results are highly favorable, we suggest to apply RSLDA for various classification problems within neuroimaging and beyond.

AB - Among the numerous methods used to analyze neuroimaging data, Linear Discriminant Analysis (LDA) is commonly applied for binary classification problems. LDAs popularity derives from its simplicity and its competitive classification performance, which has been reported for various types of neuroimaging data.Yet the standard LDA approach proves less than optimal for binary classification problems when additional label information (i.e. subclass labels) is present. Subclass labels allow to model structure in the data, which can be used to facilitate the classification task. In this paper, we illustrate how neuroimaging data exhibit subclass labels that may contain valuable information. We also show that the standard LDA classifier is unable to exploit subclass labels. We introduce a novel method that allows subclass labels to be incorporated efficiently into the classifier. The novel method, which we call Relevance Subclass LDA (RSLDA), computes an individual classification hyperplane for each subclass. It is based on regularized estimators of the subclass mean and uses other subclasses as regularization targets. We demonstrate the applicability and performance of our method on data drawn from two different neuroimaging modalities: (I) EEG data from brain-computer interfacing with event-related potentials, and (II) fMRI data in response to different levels of visual motion. We show that RSLDA outperforms the standard LDA approach for both types of datasets. These findings illustrate the benefits of exploiting subclass structure in neuroimaging data. Finally, we show that our classifier also outputs regularization profiles, enabling researchers to interpret the subclass structure in a meaningful way.RSLDA therefore yields increased classification accuracy as well as a better interpretation of neuroimaging data. Since both results are highly favorable, we suggest to apply RSLDA for various classification problems within neuroimaging and beyond.

KW - BCI

KW - Data-driven clustering

KW - EEG

KW - ERP

KW - FMRI

KW - Linear classifier

KW - Pattern classification

KW - Regularization profile

KW - Searchlight

KW - Shrinkage

KW - Single-trial classification

KW - Subclass

UR - http://www.scopus.com/inward/record.url?scp=84943621487&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943621487&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2015.09.031

DO - 10.1016/j.neuroimage.2015.09.031

M3 - Article

VL - 124

SP - 740

EP - 751

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -