Rene Gutierrez, Aaron Wolfe Sceffler and Rajarshi Guhaniyogi
05/31/2022 09:19 AM
Statistics
Clinical researchers often collect multiple images from separate modalities (sources) to investigate fundamental questions of human health that are inadequately explained by considering one image source at a time. Viewing the collection of images as multiple objects, the successful integration of multi-object data produces a sum of information greater than the individual parts, but this integration can be challenging due to the complexity induced through different topological structure of the objects. This article focuses on a multi-modal imaging application where structural/anatomical information from grey matter (GM) and brain connectivity information in the form of a brain connectome network from functional magnetic resonance imaging (fMRI) are available for a limited number of subjects with different degrees of primary progressive aphasia (PPA), a neurodegenerative disorder (NDs) measured through a score on language loss. The clinical/scientific goal in this application becomes the identification of brain regions significantly related to the language score to gain insight into ND pathways. This article develops a flexible Bayesian regression framework exploiting network information of the brain connectome, while leveraging linkages among connectome network and anatomical information from GM to draw inference on brain regions significantly related to the language score. The principled Bayesian framework allows precise characterization of the uncertainty in ascertaining a region being actively related to the language score. Our framework is implemented using an efficient Markov Chain Monte Carlo algorithm. Empirical results with simulated data illustrate substantial inferential gains of the proposed framework over its popular competitors. Our framework yields new insights into the relationship of brain regions with PPA, also providing the uncertainty associated with the scientific findings.
