Variational Bayesian mixed-effects inference for classification studies

• Published in: NeuroImage (Orlando, Fla.) Vol. 76; pp. 345 - 361
• Main Authors: Brodersen, Kay H., Daunizeau, Jean, Mathys, Christoph, Chumbley, Justin R., Buhmann, Joachim M., Stephan, Klaas E.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.08.2013; Elsevier; Elsevier Limited
• Subjects: Accuracy; Algorithms; Balanced accuracy; Bayes Theorem; Bayesian inference; Behavior; Biological and medical sciences; Brain - physiology; Brain Mapping - methods; Brain research; Classification; Estimates; Fixed effects; Fundamental and applied biological sciences. Psychology; Group studies; Humans; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging; Medical imaging; Models, Neurological; Normal-binomial; Performance evaluation; Population; Random effects; Software; Studies; Variational Bayes; Vertebrates: nervous system and sense organs; Balanced accuracy; Variational Bayes; Normal-binomial; Group studies; Fixed effects; Random effects; Bayesian inference
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2013.03.008

Multivariate classification algorithms are powerful tools for predicting cognitive or pathophysiological states from neuroimaging data. Assessing the utility of a classifier in application domains such as cognitive neuroscience, brain–computer interfaces, or clinical diagnostics necessitates inference on classification performance at more than one level, i.e., both in individual subjects and in the population from which these subjects were sampled. Such inference requires models that explicitly account for both fixed-effects (within-subjects) and random-effects (between-subjects) variance components. While models of this sort are standard in mass-univariate analyses of fMRI data, they have not yet received much attention in multivariate classification studies of neuroimaging data, presumably because of the high computational costs they entail. This paper extends a recently developed hierarchical model for mixed-effects inference in multivariate classification studies and introduces an efficient variational Bayes approach to inference. Using both synthetic and empirical fMRI data, we show that this approach is equally simple to use as, yet more powerful than, a conventional t-test on subject-specific sample accuracies, and computationally much more efficient than previous sampling algorithms and permutation tests. Our approach is independent of the type of underlying classifier and thus widely applicable. The present framework may help establish mixed-effects inference as a future standard for classification group analyses. [Display omitted] •Random-/mixed-effects inference is standard in univariate neuroimaging analyses.•But it has not yet received much attention in multivariate classification studies.•We propose a novel, statistically powerful, and efficient variational Bayes method.•Our approach replaces previous, computationally expensive MCMC algorithms.•We provide MATLAB and R implementations for use in future classification studies.