Removing facial features from structural MRI images biases visual quality assessment.

A critical step before data-sharing of human neuroimaging is removing facial features to protect individuals' privacy. However, not only does this process redact identifiable information about individuals, but it also removes non-identifiable information. This introduces undesired variability i...

Full description

Saved in:
Bibliographic Details
Main Authors: Céline Provins, Élodie Savary, Thomas Sanchez, Emeline Mullier, Jaime Barranco, Elda Fischi-Gómez, Yasser Alemán-Gómez, Jonas Richiardi, Russell A Poldrack, Patric Hagmann, Oscar Esteban
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-04-01
Series:PLoS Biology
Online Access:https://doi.org/10.1371/journal.pbio.3003149
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850129047231660032
author Céline Provins
Élodie Savary
Thomas Sanchez
Emeline Mullier
Jaime Barranco
Elda Fischi-Gómez
Yasser Alemán-Gómez
Jonas Richiardi
Russell A Poldrack
Patric Hagmann
Oscar Esteban
author_facet Céline Provins
Élodie Savary
Thomas Sanchez
Emeline Mullier
Jaime Barranco
Elda Fischi-Gómez
Yasser Alemán-Gómez
Jonas Richiardi
Russell A Poldrack
Patric Hagmann
Oscar Esteban
author_sort Céline Provins
collection DOAJ
description A critical step before data-sharing of human neuroimaging is removing facial features to protect individuals' privacy. However, not only does this process redact identifiable information about individuals, but it also removes non-identifiable information. This introduces undesired variability into downstream analysis and interpretation. This registered report investigated the degree to which the so-called defacing altered the quality assessment of T1-weighted images of the human brain from the openly available "IXI dataset". The effect of defacing on manual quality assessment was investigated on a single-site subset of the dataset (N = 185). By comparing two linear mixed-effects models, we determined that four trained human raters' perception of quality was significantly influenced by defacing by modeling their ratings on the same set of images in two conditions: "nondefaced" (i.e., preserving facial features) and "defaced". In addition, we investigated these biases on automated quality assessments by applying repeated-measures, multivariate ANOVA (rm-MANOVA) on the image quality metrics extracted with MRIQC on the full IXI dataset (N = 581; three acquisition sites). This study found that defacing altered the quality assessments by humans and showed that MRIQC's quality metrics were mostly insensitive to defacing.
format Article
id doaj-art-05fd09846da247faae8b8bad104db7af
institution OA Journals
issn 1544-9173
1545-7885
language English
publishDate 2025-04-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS Biology
spelling doaj-art-05fd09846da247faae8b8bad104db7af2025-08-20T02:33:07ZengPublic Library of Science (PLoS)PLoS Biology1544-91731545-78852025-04-01234e300314910.1371/journal.pbio.3003149Removing facial features from structural MRI images biases visual quality assessment.Céline ProvinsÉlodie SavaryThomas SanchezEmeline MullierJaime BarrancoElda Fischi-GómezYasser Alemán-GómezJonas RichiardiRussell A PoldrackPatric HagmannOscar EstebanA critical step before data-sharing of human neuroimaging is removing facial features to protect individuals' privacy. However, not only does this process redact identifiable information about individuals, but it also removes non-identifiable information. This introduces undesired variability into downstream analysis and interpretation. This registered report investigated the degree to which the so-called defacing altered the quality assessment of T1-weighted images of the human brain from the openly available "IXI dataset". The effect of defacing on manual quality assessment was investigated on a single-site subset of the dataset (N = 185). By comparing two linear mixed-effects models, we determined that four trained human raters' perception of quality was significantly influenced by defacing by modeling their ratings on the same set of images in two conditions: "nondefaced" (i.e., preserving facial features) and "defaced". In addition, we investigated these biases on automated quality assessments by applying repeated-measures, multivariate ANOVA (rm-MANOVA) on the image quality metrics extracted with MRIQC on the full IXI dataset (N = 581; three acquisition sites). This study found that defacing altered the quality assessments by humans and showed that MRIQC's quality metrics were mostly insensitive to defacing.https://doi.org/10.1371/journal.pbio.3003149
spellingShingle Céline Provins
Élodie Savary
Thomas Sanchez
Emeline Mullier
Jaime Barranco
Elda Fischi-Gómez
Yasser Alemán-Gómez
Jonas Richiardi
Russell A Poldrack
Patric Hagmann
Oscar Esteban
Removing facial features from structural MRI images biases visual quality assessment.
PLoS Biology
title Removing facial features from structural MRI images biases visual quality assessment.
title_full Removing facial features from structural MRI images biases visual quality assessment.
title_fullStr Removing facial features from structural MRI images biases visual quality assessment.
title_full_unstemmed Removing facial features from structural MRI images biases visual quality assessment.
title_short Removing facial features from structural MRI images biases visual quality assessment.
title_sort removing facial features from structural mri images biases visual quality assessment
url https://doi.org/10.1371/journal.pbio.3003149
work_keys_str_mv AT celineprovins removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT elodiesavary removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT thomassanchez removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT emelinemullier removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT jaimebarranco removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT eldafischigomez removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT yasseralemangomez removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT jonasrichiardi removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT russellapoldrack removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT patrichagmann removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment
AT oscaresteban removingfacialfeaturesfromstructuralmriimagesbiasesvisualqualityassessment