Amygdalar nuclei and hippocampal subfields on MRI: Test-retest reliability of automated volumetry across different MRI sites and vendors

Giulia Quattrini; Michela Pievani; Jorge Jovicich; Marco Aiello; Nuria Bargallo; Frederik Barkhof; David Bartres-Faz; Alberto Beltramello; Francesca B. Pizzini; Olivier Blin; Regis Bordet; Massimo Caulo; Manos Constantinides; Mira Didic; Antonios Drevelegas; Antonio Ferretti; Ute Fiedler; Piero Floridi; Helene Gros-Dagnac; Karl-Titus Hoffmann; Tilman Hensch; Joost P. Kuijer; Renaud Lopes; Camillo Marra; Bernhard W. Mueller; Flavio Nobili; Lucilla Parnetti; Pierre Payoux; Agnese Picco; Jean-Philippe Ranjeva; Luca Roccatagliata; Paolo M. Rossini; Marco Salvatore; Peter Schonknecht; Bjoern H. Schott; Julien Sein; Andrea Soricelli; Roberto Tarducci; Magda Tsolaki; Pieter J. Visser; Jens Wiltfang; Jill C. Richardson; Giovanni B. Frisoni; Moira Marizzoni; PharmaCog Consortium

doi:10.1016/j.neuroimage.2020.116932

Amygdalar nuclei and hippocampal subfields on MRI: Test-retest reliability of automated volumetry across different MRI sites and vendors

Giulia Quattrini^*, Michela Pievani, Jorge Jovicich, Marco Aiello, Nuria Bargallo, Frederik Barkhof, David Bartres-Faz, Alberto Beltramello, Francesca B. Pizzini, Olivier Blin, Regis Bordet, Massimo Caulo, Manos Constantinides, Mira Didic, Antonios Drevelegas, Antonio Ferretti, Ute Fiedler, Piero Floridi, Helene Gros-Dagnac, Karl-Titus HoffmannTilman Hensch, Joost P. Kuijer, Renaud Lopes, Camillo Marra, Bernhard W. Mueller, Flavio Nobili, Lucilla Parnetti, Pierre Payoux, Agnese Picco, Jean-Philippe Ranjeva, Luca Roccatagliata, Paolo M. Rossini, Marco Salvatore, Peter Schonknecht, Bjoern H. Schott, Julien Sein, Andrea Soricelli, Roberto Tarducci, Magda Tsolaki, Pieter J. Visser, Jens Wiltfang, Jill C. Richardson, Giovanni B. Frisoni, Moira Marizzoni, PharmaCog Consortium

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

14 Citations (Web of Science)

Abstract

Background: The amygdala and the hippocampus are two limbic structures that play a critical role in cognition and behavior, however their manual segmentation and that of their smaller nuclei/subfields in multicenter datasets is time consuming and difficult due to the low contrast of standard MRI. Here, we assessed the reliability of the automated segmentation of amygdalar nuclei and hippocampal subfields across sites and vendors using FreeSurfer in two independent cohorts of older and younger healthy adults.

Methods: Sixty-five healthy older (cohort 1) and 68 younger subjects (cohort 2), from the PharmaCog and CoRR consortia, underwent repeated 3D-T1 MRI (interval 1-90 days). Segmentation was performed using FreeSurfer v6.0. Reliability was assessed using volume reproducibility error (epsilon) and spatial overlapping coefficient (DICE) between test and retest session.

Results: Significant MRI site and vendor effects (p <.05) were found in a few subfields/nuclei for the epsilon, while extensive effects were found for the DICE score of most subfields/nuclei. Reliability was strongly influenced by volume, as e correlated negatively and DICE correlated positively with volume size of structures (absolute value of Spearman's r correlations >0.43, p <1.39E-36). In particular, volumes larger than 200 mm(3) (for amygdalar nuclei) and 300 mm(3) (for hippocampal subfields, except for molecular layer) had the best test-retest reproducibility (epsilon <5% and DICE > 0.80).

Conclusion: Our results support the use of volumetric measures of larger amygdalar nuclei and hippocampal subfields in multisite MRI studies. These measures could be useful for disease tracking and assessment of efficacy in drug trials.

Original language	English
Article number	116932
Number of pages	14
Journal	Neuroimage
Volume	218
DOIs	https://doi.org/10.1016/j.neuroimage.2020.116932
Publication status	Published - Sept 2020

Keywords

Amygdalar nuclei
Hippocampal subfields
Multicenter MRI study
FreeSurfer
Reliability analysis
BORDERLINE PERSONALITY-DISORDER
SAMPLE-SIZE CALCULATIONS
SURFACE-BASED ANALYSIS
ALZHEIMERS-DISEASE
FUNCTIONAL-ORGANIZATION
STRUCTURAL DIFFERENCES
MANUAL SEGMENTATION
METHODS FREESURFER
ELDERLY-PATIENTS
HUMAN BRAIN

Access to Document

10.1016/j.neuroimage.2020.116932Licence: CC BY-NC-ND

Cite this

Quattrini, G., Pievani, M., Jovicich, J., Aiello, M., Bargallo, N., Barkhof, F., Bartres-Faz, D., Beltramello, A., Pizzini, F. B., Blin, O., Bordet, R., Caulo, M., Constantinides, M., Didic, M., Drevelegas, A., Ferretti, A., Fiedler, U., Floridi, P., Gros-Dagnac, H., ... PharmaCog Consortium (2020). Amygdalar nuclei and hippocampal subfields on MRI: Test-retest reliability of automated volumetry across different MRI sites and vendors. Neuroimage, 218, [116932]. https://doi.org/10.1016/j.neuroimage.2020.116932

@article{c6c95095569f499a874d8831e2f2fd2c,

title = "Amygdalar nuclei and hippocampal subfields on MRI: Test-retest reliability of automated volumetry across different MRI sites and vendors",

abstract = "Background: The amygdala and the hippocampus are two limbic structures that play a critical role in cognition and behavior, however their manual segmentation and that of their smaller nuclei/subfields in multicenter datasets is time consuming and difficult due to the low contrast of standard MRI. Here, we assessed the reliability of the automated segmentation of amygdalar nuclei and hippocampal subfields across sites and vendors using FreeSurfer in two independent cohorts of older and younger healthy adults.Methods: Sixty-five healthy older (cohort 1) and 68 younger subjects (cohort 2), from the PharmaCog and CoRR consortia, underwent repeated 3D-T1 MRI (interval 1-90 days). Segmentation was performed using FreeSurfer v6.0. Reliability was assessed using volume reproducibility error (epsilon) and spatial overlapping coefficient (DICE) between test and retest session.Results: Significant MRI site and vendor effects (p <.05) were found in a few subfields/nuclei for the epsilon, while extensive effects were found for the DICE score of most subfields/nuclei. Reliability was strongly influenced by volume, as e correlated negatively and DICE correlated positively with volume size of structures (absolute value of Spearman's r correlations >0.43, p <1.39E-36). In particular, volumes larger than 200 mm(3) (for amygdalar nuclei) and 300 mm(3) (for hippocampal subfields, except for molecular layer) had the best test-retest reproducibility (epsilon <5% and DICE > 0.80).Conclusion: Our results support the use of volumetric measures of larger amygdalar nuclei and hippocampal subfields in multisite MRI studies. These measures could be useful for disease tracking and assessment of efficacy in drug trials.",

keywords = "Amygdalar nuclei, Hippocampal subfields, Multicenter MRI study, FreeSurfer, Reliability analysis, BORDERLINE PERSONALITY-DISORDER, SAMPLE-SIZE CALCULATIONS, SURFACE-BASED ANALYSIS, ALZHEIMERS-DISEASE, FUNCTIONAL-ORGANIZATION, STRUCTURAL DIFFERENCES, MANUAL SEGMENTATION, METHODS FREESURFER, ELDERLY-PATIENTS, HUMAN BRAIN",

author = "Giulia Quattrini and Michela Pievani and Jorge Jovicich and Marco Aiello and Nuria Bargallo and Frederik Barkhof and David Bartres-Faz and Alberto Beltramello and Pizzini, {Francesca B.} and Olivier Blin and Regis Bordet and Massimo Caulo and Manos Constantinides and Mira Didic and Antonios Drevelegas and Antonio Ferretti and Ute Fiedler and Piero Floridi and Helene Gros-Dagnac and Karl-Titus Hoffmann and Tilman Hensch and Kuijer, {Joost P.} and Renaud Lopes and Camillo Marra and Mueller, {Bernhard W.} and Flavio Nobili and Lucilla Parnetti and Pierre Payoux and Agnese Picco and Jean-Philippe Ranjeva and Luca Roccatagliata and Rossini, {Paolo M.} and Marco Salvatore and Peter Schonknecht and Schott, {Bjoern H.} and Julien Sein and Andrea Soricelli and Roberto Tarducci and Magda Tsolaki and Visser, {Pieter J.} and Jens Wiltfang and Richardson, {Jill C.} and Frisoni, {Giovanni B.} and Moira Marizzoni and {PharmaCog Consortium}",

year = "2020",

month = sep,

doi = "10.1016/j.neuroimage.2020.116932",

language = "English",

volume = "218",

journal = "Neuroimage",

issn = "1053-8119",

publisher = "Elsevier Science",

}

Quattrini, G, Pievani, M, Jovicich, J, Aiello, M, Bargallo, N, Barkhof, F, Bartres-Faz, D, Beltramello, A, Pizzini, FB, Blin, O, Bordet, R, Caulo, M, Constantinides, M, Didic, M, Drevelegas, A, Ferretti, A, Fiedler, U, Floridi, P, Gros-Dagnac, H, Hoffmann, K-T, Hensch, T, Kuijer, JP, Lopes, R, Marra, C, Mueller, BW, Nobili, F, Parnetti, L, Payoux, P, Picco, A, Ranjeva, J-P, Roccatagliata, L, Rossini, PM, Salvatore, M, Schonknecht, P, Schott, BH, Sein, J, Soricelli, A, Tarducci, R, Tsolaki, M, Visser, PJ, Wiltfang, J, Richardson, JC, Frisoni, GB, Marizzoni, M & PharmaCog Consortium 2020, 'Amygdalar nuclei and hippocampal subfields on MRI: Test-retest reliability of automated volumetry across different MRI sites and vendors', Neuroimage, vol. 218, 116932. https://doi.org/10.1016/j.neuroimage.2020.116932

TY - JOUR

T1 - Amygdalar nuclei and hippocampal subfields on MRI

T2 - Test-retest reliability of automated volumetry across different MRI sites and vendors

AU - Quattrini, Giulia

AU - Pievani, Michela

AU - Jovicich, Jorge

AU - Aiello, Marco

AU - Bargallo, Nuria

AU - Barkhof, Frederik

AU - Bartres-Faz, David

AU - Beltramello, Alberto

AU - Pizzini, Francesca B.

AU - Blin, Olivier

AU - Bordet, Regis

AU - Caulo, Massimo

AU - Constantinides, Manos

AU - Didic, Mira

AU - Drevelegas, Antonios

AU - Ferretti, Antonio

AU - Fiedler, Ute

AU - Floridi, Piero

AU - Gros-Dagnac, Helene

AU - Hoffmann, Karl-Titus

AU - Hensch, Tilman

AU - Kuijer, Joost P.

AU - Lopes, Renaud

AU - Marra, Camillo

AU - Mueller, Bernhard W.

AU - Nobili, Flavio

AU - Parnetti, Lucilla

AU - Payoux, Pierre

AU - Picco, Agnese

AU - Ranjeva, Jean-Philippe

AU - Roccatagliata, Luca

AU - Rossini, Paolo M.

AU - Salvatore, Marco

AU - Schonknecht, Peter

AU - Schott, Bjoern H.

AU - Sein, Julien

AU - Soricelli, Andrea

AU - Tarducci, Roberto

AU - Tsolaki, Magda

AU - Visser, Pieter J.

AU - Wiltfang, Jens

AU - Richardson, Jill C.

AU - Frisoni, Giovanni B.

AU - Marizzoni, Moira

AU - PharmaCog Consortium

PY - 2020/9

Y1 - 2020/9

N2 - Background: The amygdala and the hippocampus are two limbic structures that play a critical role in cognition and behavior, however their manual segmentation and that of their smaller nuclei/subfields in multicenter datasets is time consuming and difficult due to the low contrast of standard MRI. Here, we assessed the reliability of the automated segmentation of amygdalar nuclei and hippocampal subfields across sites and vendors using FreeSurfer in two independent cohorts of older and younger healthy adults.Methods: Sixty-five healthy older (cohort 1) and 68 younger subjects (cohort 2), from the PharmaCog and CoRR consortia, underwent repeated 3D-T1 MRI (interval 1-90 days). Segmentation was performed using FreeSurfer v6.0. Reliability was assessed using volume reproducibility error (epsilon) and spatial overlapping coefficient (DICE) between test and retest session.Results: Significant MRI site and vendor effects (p <.05) were found in a few subfields/nuclei for the epsilon, while extensive effects were found for the DICE score of most subfields/nuclei. Reliability was strongly influenced by volume, as e correlated negatively and DICE correlated positively with volume size of structures (absolute value of Spearman's r correlations >0.43, p <1.39E-36). In particular, volumes larger than 200 mm(3) (for amygdalar nuclei) and 300 mm(3) (for hippocampal subfields, except for molecular layer) had the best test-retest reproducibility (epsilon <5% and DICE > 0.80).Conclusion: Our results support the use of volumetric measures of larger amygdalar nuclei and hippocampal subfields in multisite MRI studies. These measures could be useful for disease tracking and assessment of efficacy in drug trials.

AB - Background: The amygdala and the hippocampus are two limbic structures that play a critical role in cognition and behavior, however their manual segmentation and that of their smaller nuclei/subfields in multicenter datasets is time consuming and difficult due to the low contrast of standard MRI. Here, we assessed the reliability of the automated segmentation of amygdalar nuclei and hippocampal subfields across sites and vendors using FreeSurfer in two independent cohorts of older and younger healthy adults.Methods: Sixty-five healthy older (cohort 1) and 68 younger subjects (cohort 2), from the PharmaCog and CoRR consortia, underwent repeated 3D-T1 MRI (interval 1-90 days). Segmentation was performed using FreeSurfer v6.0. Reliability was assessed using volume reproducibility error (epsilon) and spatial overlapping coefficient (DICE) between test and retest session.Results: Significant MRI site and vendor effects (p <.05) were found in a few subfields/nuclei for the epsilon, while extensive effects were found for the DICE score of most subfields/nuclei. Reliability was strongly influenced by volume, as e correlated negatively and DICE correlated positively with volume size of structures (absolute value of Spearman's r correlations >0.43, p <1.39E-36). In particular, volumes larger than 200 mm(3) (for amygdalar nuclei) and 300 mm(3) (for hippocampal subfields, except for molecular layer) had the best test-retest reproducibility (epsilon <5% and DICE > 0.80).Conclusion: Our results support the use of volumetric measures of larger amygdalar nuclei and hippocampal subfields in multisite MRI studies. These measures could be useful for disease tracking and assessment of efficacy in drug trials.

KW - Amygdalar nuclei

KW - Hippocampal subfields

KW - Multicenter MRI study

KW - FreeSurfer

KW - Reliability analysis

KW - BORDERLINE PERSONALITY-DISORDER

KW - SAMPLE-SIZE CALCULATIONS

KW - SURFACE-BASED ANALYSIS

KW - ALZHEIMERS-DISEASE

KW - FUNCTIONAL-ORGANIZATION

KW - STRUCTURAL DIFFERENCES

KW - MANUAL SEGMENTATION

KW - METHODS FREESURFER

KW - ELDERLY-PATIENTS

KW - HUMAN BRAIN

U2 - 10.1016/j.neuroimage.2020.116932

DO - 10.1016/j.neuroimage.2020.116932

M3 - Article

C2 - 32416226

SN - 1053-8119

VL - 218

JO - Neuroimage

JF - Neuroimage

M1 - 116932

ER -