The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Standard

The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data. / Nørgaard, Martin; Greve, Douglas N.; Svarer, Claus; Strother, Stephen C.; Knudsen, Gitte M.; Ganz, Melanie.

2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018. IEEE, 2018. 8423962.

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Harvard

Nørgaard, M, Greve, DN, Svarer, C, Strother, SC, Knudsen, GM & Ganz, M 2018, The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data. i 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018., 8423962, IEEE, 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018, Singapore, Singapore, 12/06/2018. https://doi.org/10.1109/PRNI.2018.8423962

APA

Nørgaard, M., Greve, D. N., Svarer, C., Strother, S. C., Knudsen, G. M., & Ganz, M. (2018). The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data. I 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018 [8423962] IEEE. https://doi.org/10.1109/PRNI.2018.8423962

Vancouver

Nørgaard M, Greve DN, Svarer C, Strother SC, Knudsen GM, Ganz M. The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data. I 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018. IEEE. 2018. 8423962 https://doi.org/10.1109/PRNI.2018.8423962

Author

Nørgaard, Martin ; Greve, Douglas N. ; Svarer, Claus ; Strother, Stephen C. ; Knudsen, Gitte M. ; Ganz, Melanie. / The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data. 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018. IEEE, 2018.

Bibtex

@inproceedings{4048db38adb64c218911db54448a69c1,
title = "The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data",
abstract = "It has long been recognized that the data preprocessing chain is a critical part of a neuroimaging experiment. In this work we evaluate the impact of preprocessing choices in univariate and multivariate analyses of Positron Emission Tomography (PET) data. Thirty healthy participants were scanned twice in a High-Resolution Research Tomography PET scanner with the serotonin transporter (5-HTT) radioligand [11 C]DASB. Binding potentials (BPND) from 14 brain regions are quantified with 384 different preprocessing choices. A univariate paired t-Test is applied to each region and for each preprocessing choice, and corrected for multiple comparisons using FDR within each pipeline. Additionally, a multivariate Linear Discriminant Analysis (LDA) model is used to discriminate test and retest BPND, and the model performance is evaluated using a repeated cross-validation framework with permutations. The univariate analysis revealed several significant differences in 5-HTT BPND across brain regions, depending on the preprocessing choice. The classification accuracy of the multivariate LDA model varied from 37{\%} to 70{\%} depending on the choice of preprocessing, and could reasonably be modeled with a normal distribution centered at 51{\%} accuracy. In spite of correcting for multiple comparisons, the univariate model with varying preprocessing choices is more likely to generate false-positive results compared to a simple multivariate analysis model evaluated with cross-validation and permutations.",
author = "Martin N{\o}rgaard and Greve, {Douglas N.} and Claus Svarer and Strother, {Stephen C.} and Knudsen, {Gitte M.} and Melanie Ganz",
year = "2018",
doi = "10.1109/PRNI.2018.8423962",
language = "English",
isbn = "9781538668597",
booktitle = "2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018",
publisher = "IEEE",
note = "2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018 ; Conference date: 12-06-2018 Through 14-06-2018",

}

RIS

TY - GEN

T1 - The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data

AU - Nørgaard, Martin

AU - Greve, Douglas N.

AU - Svarer, Claus

AU - Strother, Stephen C.

AU - Knudsen, Gitte M.

AU - Ganz, Melanie

PY - 2018

Y1 - 2018

N2 - It has long been recognized that the data preprocessing chain is a critical part of a neuroimaging experiment. In this work we evaluate the impact of preprocessing choices in univariate and multivariate analyses of Positron Emission Tomography (PET) data. Thirty healthy participants were scanned twice in a High-Resolution Research Tomography PET scanner with the serotonin transporter (5-HTT) radioligand [11 C]DASB. Binding potentials (BPND) from 14 brain regions are quantified with 384 different preprocessing choices. A univariate paired t-Test is applied to each region and for each preprocessing choice, and corrected for multiple comparisons using FDR within each pipeline. Additionally, a multivariate Linear Discriminant Analysis (LDA) model is used to discriminate test and retest BPND, and the model performance is evaluated using a repeated cross-validation framework with permutations. The univariate analysis revealed several significant differences in 5-HTT BPND across brain regions, depending on the preprocessing choice. The classification accuracy of the multivariate LDA model varied from 37% to 70% depending on the choice of preprocessing, and could reasonably be modeled with a normal distribution centered at 51% accuracy. In spite of correcting for multiple comparisons, the univariate model with varying preprocessing choices is more likely to generate false-positive results compared to a simple multivariate analysis model evaluated with cross-validation and permutations.

AB - It has long been recognized that the data preprocessing chain is a critical part of a neuroimaging experiment. In this work we evaluate the impact of preprocessing choices in univariate and multivariate analyses of Positron Emission Tomography (PET) data. Thirty healthy participants were scanned twice in a High-Resolution Research Tomography PET scanner with the serotonin transporter (5-HTT) radioligand [11 C]DASB. Binding potentials (BPND) from 14 brain regions are quantified with 384 different preprocessing choices. A univariate paired t-Test is applied to each region and for each preprocessing choice, and corrected for multiple comparisons using FDR within each pipeline. Additionally, a multivariate Linear Discriminant Analysis (LDA) model is used to discriminate test and retest BPND, and the model performance is evaluated using a repeated cross-validation framework with permutations. The univariate analysis revealed several significant differences in 5-HTT BPND across brain regions, depending on the preprocessing choice. The classification accuracy of the multivariate LDA model varied from 37% to 70% depending on the choice of preprocessing, and could reasonably be modeled with a normal distribution centered at 51% accuracy. In spite of correcting for multiple comparisons, the univariate model with varying preprocessing choices is more likely to generate false-positive results compared to a simple multivariate analysis model evaluated with cross-validation and permutations.

UR - http://www.scopus.com/inward/record.url?scp=85051565165&partnerID=8YFLogxK

U2 - 10.1109/PRNI.2018.8423962

DO - 10.1109/PRNI.2018.8423962

M3 - Article in proceedings

AN - SCOPUS:85051565165

SN - 9781538668597

BT - 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018

PB - IEEE

T2 - 2018 International Workshop on Pattern Recognition in Neuroimaging, PRNI 2018

Y2 - 12 June 2018 through 14 June 2018

ER -

ID: 203671902