TY - JOUR
T1 - Analyzing neuroimaging data with subclasses
T2 - A shrinkage approach
AU - Höhne, Johannes
AU - Bartz, Daniel
AU - Hebart, Martin N.
AU - Müller, Klaus Robert
AU - Blankertz, Benjamin
N1 - Funding Information:
Johannes Höhne gratefully acknowledges funding by European Union ( FP7-609593 and FP7-224631 ). Klaus-Robert Muller gratefully acknowledges funding by BMBF Big Data Centre ( 01 IS 14013 A ) and the BK21 program funded by Korean National Research Foundation grant (No. 2012-005741 ). Benjamin Blankertz gratefully acknowledges funding by BMBF Grant Nos. 16SV5839 and 01GQ0850 and European Union ( FP7-611570 ). This paper reflects only the authors' views. Funding agencies are not liable for any use that may be made of the information contained herein. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Publisher Copyright:
© 2015 Elsevier Inc.
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
U2 - 10.1016/j.neuroimage.2015.09.031
DO - 10.1016/j.neuroimage.2015.09.031
M3 - Article
C2 - 26407815
AN - SCOPUS:84943621487
SN - 1053-8119
VL - 124
SP - 740
EP - 751
JO - NeuroImage
JF - NeuroImage
ER -