Smooth Scalar-on-Image Regression via Spatial Bayesian Variable Selection

• Published in: Journal of computational and graphical statistics Vol. 23; no. 1; pp. 46 - 64
• Main Authors: Goldsmith, Jeff, Huang, Lei, Crainiceanu, Ciprian M.
• Format: Journal Article
• Language: English
• Published: United States Taylor & Francis 2014; American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America; Taylor & Francis Ltd
• Subjects: Bayesian Methodology; Binary Markov random field; Cognition; Corpus callosum; Datasets; Estimation methods; Gaussian Markov random field; Graphics; Linear regression; Markov chain Monte Carlo; Mathematical models; Medical imaging; Multiple sclerosis; Probability distribution; Regression analysis; Regression coefficients; Simulations; Statistical inference; Studies; Binary Markov Random Field; Markov Chain Monte Carlo; Gaussian Markov Random Field
• ISSN: 1061-8600; 1537-2715
• DOI: 10.1080/10618600.2012.743437

We develop scalar-on-image regression models when images are registered multidimensional manifolds. We propose a fast and scalable Bayes' inferential procedure to estimate the image coefficient. The central idea is the combination of an Ising prior distribution, which controls a latent binary indicator map, and an intrinsic Gaussian Markov random field, which controls the smoothness of the nonzero coefficients. The model is fit using a single-site Gibbs sampler, which allows fitting within minutes for hundreds of subjects with predictor images containing thousands of locations. The code is simple and is provided in the online Appendix (see the "Supplementary Materials" section). We apply this method to a neuroimaging study where cognitive outcomes are regressed on measures of white-matter microstructure at every voxel of the corpus callosum for hundreds of subjects.