Understanding the impact of preprocessing pipelines on neuroimaging cortical surface analyses.
cortical thickness
neuroimaging
preprocessing pipelines
reproducibility
Journal
GigaScience
ISSN: 2047-217X
Titre abrégé: Gigascience
Pays: United States
ID NLM: 101596872
Informations de publication
Date de publication:
22 01 2021
22 01 2021
Historique:
received:
04
08
2020
revised:
01
11
2020
entrez:
22
1
2021
pubmed:
23
1
2021
medline:
26
10
2021
Statut:
ppublish
Résumé
The choice of preprocessing pipeline introduces variability in neuroimaging analyses that affects the reproducibility of scientific findings. Features derived from structural and functional MRI data are sensitive to the algorithmic or parametric differences of preprocessing tasks, such as image normalization, registration, and segmentation to name a few. Therefore it is critical to understand and potentially mitigate the cumulative biases of pipelines in order to distinguish biological effects from methodological variance. Here we use an open structural MRI dataset (ABIDE), supplemented with the Human Connectome Project, to highlight the impact of pipeline selection on cortical thickness measures. Specifically, we investigate the effect of (i) software tool (e.g., ANTS, CIVET, FreeSurfer), (ii) cortical parcellation (Desikan-Killiany-Tourville, Destrieux, Glasser), and (iii) quality control procedure (manual, automatic). We divide our statistical analyses by (i) method type, i.e., task-free (unsupervised) versus task-driven (supervised); and (ii) inference objective, i.e., neurobiological group differences versus individual prediction. Results show that software, parcellation, and quality control significantly affect task-driven neurobiological inference. Additionally, software selection strongly affects neurobiological (i.e. group) and individual task-free analyses, and quality control alters the performance for the individual-centric prediction tasks. This comparative performance evaluation partially explains the source of inconsistencies in neuroimaging findings. Furthermore, it underscores the need for more rigorous scientific workflows and accessible informatics resources to replicate and compare preprocessing pipelines to address the compounding problem of reproducibility in the age of large-scale, data-driven computational neuroscience.
Sections du résumé
BACKGROUND
The choice of preprocessing pipeline introduces variability in neuroimaging analyses that affects the reproducibility of scientific findings. Features derived from structural and functional MRI data are sensitive to the algorithmic or parametric differences of preprocessing tasks, such as image normalization, registration, and segmentation to name a few. Therefore it is critical to understand and potentially mitigate the cumulative biases of pipelines in order to distinguish biological effects from methodological variance.
METHODS
Here we use an open structural MRI dataset (ABIDE), supplemented with the Human Connectome Project, to highlight the impact of pipeline selection on cortical thickness measures. Specifically, we investigate the effect of (i) software tool (e.g., ANTS, CIVET, FreeSurfer), (ii) cortical parcellation (Desikan-Killiany-Tourville, Destrieux, Glasser), and (iii) quality control procedure (manual, automatic). We divide our statistical analyses by (i) method type, i.e., task-free (unsupervised) versus task-driven (supervised); and (ii) inference objective, i.e., neurobiological group differences versus individual prediction.
RESULTS
Results show that software, parcellation, and quality control significantly affect task-driven neurobiological inference. Additionally, software selection strongly affects neurobiological (i.e. group) and individual task-free analyses, and quality control alters the performance for the individual-centric prediction tasks.
CONCLUSIONS
This comparative performance evaluation partially explains the source of inconsistencies in neuroimaging findings. Furthermore, it underscores the need for more rigorous scientific workflows and accessible informatics resources to replicate and compare preprocessing pipelines to address the compounding problem of reproducibility in the age of large-scale, data-driven computational neuroscience.
Identifiants
pubmed: 33481004
pii: 6106556
doi: 10.1093/gigascience/giaa155
pmc: PMC7821710
pii:
doi:
Banques de données
figshare
['10.6084/m9.figshare.3498446.v2']
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Subventions
Organisme : NIMH NIH HHS
ID : R01 MH096906
Pays : United States
Organisme : NIMH NIH HHS
ID : RF1 MH120021
Pays : United States
Organisme : NIBIB NIH HHS
ID : P41 EB019936
Pays : United States
Organisme : NIMH NIH HHS
ID : R01 MH083320
Pays : United States
Informations de copyright
© The Author(s) 2021. Published by Oxford University Press GigaScience.
Références
J Neurosci. 2004 Sep 22;24(38):8223-31
pubmed: 15385605
Sci Rep. 2020 Jul 6;10(1):11067
pubmed: 32632150
Front Neuroinform. 2011 Aug 22;5:13
pubmed: 21897815
Neuroimage. 2014 Oct 1;99:166-79
pubmed: 24879923
Front Neuroinform. 2019 Feb 07;13:1
pubmed: 30792636
Nature. 2016 May 25;533(7604):452-4
pubmed: 27225100
Neuroimage. 1999 Feb;9(2):179-94
pubmed: 9931268
Proc Natl Acad Sci U S A. 2018 Mar 13;115(11):2628-2631
pubmed: 29531051
Transl Psychiatry. 2018 Nov 16;8(1):249
pubmed: 30446637
Neuroinformatics. 2015 Jan;13(1):31-46
pubmed: 25048627
Psychol Med. 2018 Mar;48(4):654-668
pubmed: 28745267
Neuroimage. 2009 Apr 15;45(3):867-79
pubmed: 19150502
Nature. 2016 Aug 11;536(7615):171-178
pubmed: 27437579
Nat Hum Behav. 2018 Jan;2(1):6-10
pubmed: 30980045
Front Neurosci. 2012 Oct 11;6:149
pubmed: 23087605
Neuroimage. 2010 Oct 15;53(1):1-15
pubmed: 20547229
Nat Rev Neurosci. 2013 May;14(5):365-76
pubmed: 23571845
Cereb Cortex. 2017 Mar 1;27(3):1721-1731
pubmed: 28334080
IEEE Trans Med Imaging. 1998 Feb;17(1):87-97
pubmed: 9617910
PLoS Med. 2005 Aug;2(8):e124
pubmed: 16060722
J Neural Transm (Vienna). 2014 Sep;121(9):1157-70
pubmed: 24752753
Netw Neurosci. 2020 Mar 01;4(1):274-291
pubmed: 32181419
Gigascience. 2021 Jan 22;10(1):
pubmed: 33481004
Mol Psychiatry. 2020 Mar;25(3):614-628
pubmed: 31028290
F1000Res. 2017 Feb 10;6:124
pubmed: 28781753
Netw Neurosci. 2019 Feb 01;3(2):344-362
pubmed: 30793086
Neuroimage. 2015 May 1;111:350-9
pubmed: 25731999
Neuroimage. 2013 Oct 15;80:62-79
pubmed: 23684880
Eur J Neurosci. 2018 Mar;47(5):399-416
pubmed: 29359873
Proc Natl Acad Sci U S A. 2000 Sep 26;97(20):11050-5
pubmed: 10984517
Hum Brain Mapp. 2019 Aug 1;40(11):3362-3384
pubmed: 31050106
Mol Psychiatry. 2014 Jun;19(6):659-67
pubmed: 23774715
J Comput Assist Tomogr. 1994 Mar-Apr;18(2):192-205
pubmed: 8126267
J Comput Neurosci. 2018 Dec;45(3):163-172
pubmed: 30377880
Neuroimage. 2006 Jul 1;31(3):968-80
pubmed: 16530430
Front Neurosci. 2012 Dec 05;6:171
pubmed: 23227001
MNI Open Res. 2019;2:
pubmed: 31157322
IEEE Trans Med Imaging. 2010 Jun;29(6):1310-20
pubmed: 20378467
Proc Natl Acad Sci U S A. 2016 Jul 12;113(28):7900-5
pubmed: 27357684
Psychol Sci. 2020 Jul;31(7):792-806
pubmed: 32489141
Front Neuroinform. 2014 May 21;8:54
pubmed: 24904400
Eur Child Adolesc Psychiatry. 2005 May;14(3):138-44
pubmed: 15959659
J Neurophysiol. 2011 Sep;106(3):1125-65
pubmed: 21653723
Neuroimage. 2012 Aug 15;62(2):774-81
pubmed: 22248573