Using the General Linear Model to Improve Performance in fNIRS Single Trial Analysis and Classification: A Perspective

dc.contributor.authorvon Lühmann, Alexander
dc.contributor.authorOrtega-Martinez, Antonio
dc.contributor.authorBoas, David A.
dc.contributor.authorYücel, Meryem Ayşe
dc.date.accessioned2020-04-27T15:57:11Z
dc.date.available2020-04-27T15:57:11Z
dc.date.issued2020-02-18
dc.date.updated2020-02-18T07:27:08Z
dc.description.abstractWithin a decade, single trial analysis of functional Near Infrared Spectroscopy (fNIRS) signals has gained significant momentum, and fNIRS joined the set of modalities frequently used for active and passive Brain Computer Interfaces (BCI). A great variety of methods for feature extraction and classification have been explored using state-of-the-art Machine Learning methods. In contrast, signal preprocessing and cleaning pipelines for fNIRS often follow simple recipes and so far rarely incorporate the available state-of-the-art in adjacent fields. In neuroscience, where fMRI and fNIRS are established neuroimaging tools, evoked hemodynamic brain activity is typically estimated across multiple trials using a General Linear Model (GLM). With the help of the GLM, subject, channel, and task specific evoked hemodynamic responses are estimated, and the evoked brain activity is more robustly separated from systemic physiological interference using independent measures of nuisance regressors, such as short-separation fNIRS measurements. When correctly applied in single trial analysis, e.g., in BCI, this approach can significantly enhance contrast to noise ratio of the brain signal, improve feature separability and ultimately lead to better classification accuracy. In this manuscript, we provide a brief introduction into the GLM and show how to incorporate it into a typical BCI preprocessing pipeline and cross-validation. Using a resting state fNIRS data set augmented with synthetic hemodynamic responses that provide ground truth brain activity, we compare the quality of commonly used fNIRS features for BCI that are extracted from (1) conventionally preprocessed signals, and (2) signals preprocessed with the GLM and physiological nuisance regressors. We show that the GLM-based approach can provide better single trial estimates of brain activity as well as a new feature type, i.e., the weight of the individual and channel-specific hemodynamic response function (HRF) regressor. The improved estimates yield features with higher separability, that significantly enhance accuracy in a binary classification task when compared to conventional preprocessing—on average +7.4% across subjects and feature types. We propose to adapt this well-established approach from neuroscience to the domain of single-trial analysis and preprocessing wherever the classification of evoked brain activity is of concern, for instance in BCI.en
dc.identifier.eissn1662-5161
dc.identifier.urihttps://depositonce.tu-berlin.de/handle/11303/11032
dc.identifier.urihttp://dx.doi.org/10.14279/depositonce-9920
dc.language.isoenen
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/en
dc.subject.ddc610 Medizin und Gesundheitde
dc.subject.otherfNIRSen
dc.subject.otherBCIen
dc.subject.otherGLMen
dc.subject.otherpreprocessingen
dc.subject.otherclassificationen
dc.subject.otherHRFen
dc.subject.othershort-separationen
dc.subject.othernuisance regressionen
dc.titleUsing the General Linear Model to Improve Performance in fNIRS Single Trial Analysis and Classification: A Perspectiveen
dc.typeArticleen
dc.type.versionpublishedVersionen
dcterms.bibliographicCitation.articlenumber30en
dcterms.bibliographicCitation.doi10.3389/fnhum.2020.00030en
dcterms.bibliographicCitation.journaltitleFrontiers in Human Neuroscienceen
dcterms.bibliographicCitation.originalpublishernameFrontiers Media S.A.en
dcterms.bibliographicCitation.originalpublisherplaceLausanneen
dcterms.bibliographicCitation.volume14en
tub.accessrights.dnbfreeen
tub.affiliationFak. 4 Elektrotechnik und Informatik>Inst. Softwaretechnik und Theoretische Informatik>FG Maschinelles Lernende
tub.affiliation.facultyFak. 4 Elektrotechnik und Informatikde
tub.affiliation.groupFG Maschinelles Lernende
tub.affiliation.instituteInst. Softwaretechnik und Theoretische Informatikde
tub.publisher.universityorinstitutionTechnische Universität Berlinen
Files
Original bundle
Now showing 1 - 14 of 14
Loading…
Thumbnail Image
Name:
fnhum-14-00030.pdf
Size:
16.79 MB
Format:
Adobe Portable Document Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0001.tif
Size:
133.35 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0002.tif
Size:
420.98 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0003.tif
Size:
173.06 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0004.tif
Size:
636.43 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0005.tif
Size:
245.12 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0006.tif
Size:
632.28 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0007.tif
Size:
280.07 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0008.tif
Size:
185.39 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0009.tif
Size:
337.36 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0010.tif
Size:
179.22 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0011.tif
Size:
631.25 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
fnhum-14-00030-g0012.tif
Size:
861.5 KB
Format:
Tag Image File Format
Description:
Loading…
Thumbnail Image
Name:
Table_1.XLSX
Size:
37.51 KB
Format:
Microsoft Excel XML
Description:
Collections