TY  - JOUR
AU  - Kobeleva, Xenia
AU  - Varoquaux, Gaël
AU  - Dagher, Alain
AU  - Adhikari, Mohit
AU  - Grefkes, Christian
AU  - Gilson, Matthieu
TI  - Advancing brain network models to reconcile functional neuroimaging and clinical research
JO  - NeuroImage: Clinical
VL  - 36
SN  - 2213-1582
CY  - [Amsterdam u.a.]
PB  - Elsevier
M1  - FZJ-2022-05726
SP  - 103262 -
PY  - 2022
AB  - Functional magnetic resonance imaging (fMRI) captures information on brain function beyond the anatomicalalterations that are traditionally visually examined by neuroradiologists. However, the fMRI signals are complexin addition to being noisy, so fMRI still faces limitations for clinical applications. Here we review methods thathave been proposed as potential solutions so far, namely statistical, biophysical and decoding models, with theirstrengths and weaknesses. We especially evaluate the ability of these models to directly predict clinical variablesfrom their parameters (predictability) and to extract clinically relevant information regarding biologicalmechanisms and relevant features for classification and prediction (interpretability). We then provide guidelinesfor useful applications and pitfalls of such fMRI-based models in a clinical research context, looking beyond thecurrent state of the art. In particular, we argue that the clinical relevance of fMRI calls for a new generation ofmodels for fMRI data, which combine the strengths of both biophysical and decoding models. This leads toreliable and biologically meaningful model parameters, which thus fulfills the need for simultaneous inter-pretability and predictability. In our view, this synergy is fundamental for the discovery of new pharmacologicaland interventional targets, as well as the use of models as biomarkers in neurology and psychiatry.
LB  - PUB:(DE-HGF)16
C6  - 36451365
UR  - <Go to ISI:>//WOS:000892218700001
DO  - DOI:10.1016/j.nicl.2022.103262
UR  - https://juser.fz-juelich.de/record/912553
ER  -