Format

Send to

Choose Destination
Schizophr Res. 2018 Feb;192:167-171. doi: 10.1016/j.schres.2017.05.027. Epub 2017 Aug 24.

Multisite generalizability of schizophrenia diagnosis classification based on functional brain connectivity.

Author information

1
Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal, Montréal, Québec, Canada; Centre de Recherche de l'Institut Universitaire en Santé Mentale de Montréal, Montréal, Québec, Canada; Département de Psychiatrie, Université de Montréal, Montréal, Québec, Canada. Electronic address: pierre.orban@umontreal.ca.
2
Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal, Montréal, Québec, Canada; Département d'Informatique et de Recherche Opérationnelle, Université de Montréal, Montréal, Québec, Canada.
3
Centre de Recherche de l'Institut Universitaire en Santé Mentale de Montréal, Montréal, Québec, Canada.
4
Department of Mathematics and Statistics, La Trobe University, Bundoora, Australia.
5
Centre de Recherche de l'Institut Universitaire en Santé Mentale de Montréal, Montréal, Québec, Canada; Department of Psychology, Bishop's University, Sherbrooke, Québec, Canada.
6
Centre de Recherche de l'Institut Universitaire en Santé Mentale de Montréal, Montréal, Québec, Canada; Département de Psychiatrie, Université de Montréal, Montréal, Québec, Canada; Centre Hospitalier Universitaire de Montréal, Montréal, Québec, Canada.

Abstract

Our objective was to assess the generalizability, across sites and cognitive contexts, of schizophrenia classification based on functional brain connectivity. We tested different training-test scenarios combining fMRI data from 191 schizophrenia patients and 191 matched healthy controls obtained at 6 scanning sites and under different task conditions. Diagnosis classification accuracy generalized well to a novel site and cognitive context provided data from multiple sites were used for classifier training. By contrast, lower classification accuracy was achieved when data from a single distinct site was used for training. These findings indicate that it is beneficial to use multisite data to train fMRI-based classifiers intended for large-scale use in the clinical realm.

KEYWORDS:

Classification; Cognition; Machine learning; Multisite; Schizophrenia; fMRI

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center