TY - JOUR
T1 - MRI data-driven algorithm for the diagnosis of behavioural variant frontotemporal dementia
AU - FTLDNI Investigators, GENFI Consortium
AU - Manera, Ana L.
AU - Dadar, Mahsa
AU - Van Swieten, John Cornelis
AU - Borroni, Barbara
AU - Sanchez-Valle, Raquel
AU - Moreno, Fermin
AU - Laforce, Robert
AU - Graff, Caroline
AU - Synofzik, Matthis
AU - Galimberti, Daniela
AU - Rowe, James Benedict
AU - Masellis, Mario
AU - Tartaglia, Maria Carmela
AU - Finger, Elizabeth
AU - Vandenberghe, Rik
AU - de Mendonca, Alexandre
AU - Tagliavini, Fabrizio
AU - Santana, Isabel
AU - Butler, Christopher R.
AU - Gerhard, Alex
AU - Danek, Adrian
AU - Levin, Johannes
AU - Otto, Markus
AU - Frisoni, Giovanni
AU - Ghidoni, Roberta
AU - Sorbi, Sandro
AU - Rohrer, Jonathan Daniel
AU - Ducharme, Simon
AU - Louis Collins, D.
AU - Rosen, Howard
AU - Dickerson, Bradford C.
AU - Domoto-Reilly, Kimoko
AU - Knopman, David
AU - Boeve, Bradley F.
AU - Boxer, Adam L.
AU - Kornak, John
AU - Miller, Bruce L.
AU - Seeley, William W.
AU - Gorno-Tempini, Maria Luisa
AU - McGinnis, Scott
AU - Mandelli, Maria Luisa
AU - Afonso, Sónia
AU - Almeida, Maria Rosario
AU - Anderl-Straub, Sarah
AU - Andersson, Christin
AU - Antonell, Anna
AU - Archetti, Silvana
AU - Arighi, Andrea
AU - Balasa, Mircea
AU - Rademakers, Rosa
N1 - Publisher Copyright:
© Author(s) (or their employer(s)) 2021. No commercial re-use. See rights and permissions. Published by BMJ.
PY - 2021/6/1
Y1 - 2021/6/1
N2 - Introduction Structural brain imaging is paramount for the diagnosis of behavioural variant of frontotemporal dementia (bvFTD), but it has low sensitivity leading to erroneous or late diagnosis. Methods A total of 515 subjects from two different bvFTD cohorts (training and independent validation cohorts) were used to perform voxel-wise morphometric analysis to identify regions with significant differences between bvFTD and controls. A random forest classifier was used to individually predict bvFTD from deformation-based morphometry differences in isolation and together with semantic fluency. Tenfold cross validation was used to assess the performance of the classifier within the training cohort. A second held-out cohort of genetically confirmed bvFTD cases was used for additional validation. Results Average 10-fold cross-validation accuracy was 89% (82% sensitivity, 93% specificity) using only MRI and 94% (89% sensitivity, 98% specificity) with the addition of semantic fluency. In the separate validation cohort of definite bvFTD, accuracy was 88% (81% sensitivity, 92% specificity) with MRI and 91% (79% sensitivity, 96% specificity) with added semantic fluency scores. Conclusion Our results show that structural MRI and semantic fluency can accurately predict bvFTD at the individual subject level within a completely independent validation cohort coming from a different and independent database.
AB - Introduction Structural brain imaging is paramount for the diagnosis of behavioural variant of frontotemporal dementia (bvFTD), but it has low sensitivity leading to erroneous or late diagnosis. Methods A total of 515 subjects from two different bvFTD cohorts (training and independent validation cohorts) were used to perform voxel-wise morphometric analysis to identify regions with significant differences between bvFTD and controls. A random forest classifier was used to individually predict bvFTD from deformation-based morphometry differences in isolation and together with semantic fluency. Tenfold cross validation was used to assess the performance of the classifier within the training cohort. A second held-out cohort of genetically confirmed bvFTD cases was used for additional validation. Results Average 10-fold cross-validation accuracy was 89% (82% sensitivity, 93% specificity) using only MRI and 94% (89% sensitivity, 98% specificity) with the addition of semantic fluency. In the separate validation cohort of definite bvFTD, accuracy was 88% (81% sensitivity, 92% specificity) with MRI and 91% (79% sensitivity, 96% specificity) with added semantic fluency scores. Conclusion Our results show that structural MRI and semantic fluency can accurately predict bvFTD at the individual subject level within a completely independent validation cohort coming from a different and independent database.
UR - http://www.scopus.com/inward/record.url?scp=85102678202&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85102678202&partnerID=8YFLogxK
U2 - 10.1136/jnnp-2020-324106
DO - 10.1136/jnnp-2020-324106
M3 - Article
C2 - 33722819
AN - SCOPUS:85102678202
SN - 0022-3050
VL - 92
SP - 608
EP - 616
JO - Journal of Neurology, Neurosurgery and Psychiatry
JF - Journal of Neurology, Neurosurgery and Psychiatry
IS - 6
ER -