2017/05/14 by Shinjini Kundu, Soheil Kolouri, Kundu, Shinjini +9
Biochemistry, Genetics and Molecular Biology · Mathematics · Medicine · Neuroscience · #Advanced Neuroimaging Techniques and Applications #Cell Image Analysis Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Functional Brain Connectivity Studies #Morphological variations and asymmetry
paper · pdf · doi:10.48550/arxiv.1705.04919
openalex publication_date 2017/05/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Disease in the brain is often associated with subtle, spatially diffuse, or\ncomplex tissue changes that may lie beneath the level of gross visual\ninspection, even on magnetic resonance imaging (MRI). Unfortunately, current\ncomputer-assisted approaches that examine pre-specified features, whether\nanatomically-defined (i.e. thalamic volume, cortical thickness) or based on\npixelwise comparison (i.e. deformation-based methods), are prone to missing a\nvast array of physical changes that are not well-encapsulated by these metrics.\nIn this paper, we have developed a technique for automated pattern analysis\nthat can fully determine the relationship between brain structure and\nobservable phenotype without requiring any a priori features. Our technique,\ncalled transport-based morphometry (TBM), is an image transformation that maps\nbrain images losslessly to a domain where they become much more separable. The\nnew approach is validated on structural brain images of healthy older adult\nsubjects where even linear models for discrimination, regression, and blind\nsource separation enable TBM to independently discover the characteristic\nchanges of aging and highlight potential mechanisms by which aerobic fitness\nmay mediate brain health later in life. TBM is a generative approach that can\nprovide visualization of physically meaningful shifts in tissue distribution\nthrough inverse transformation. The proposed framework is a powerful technique\nthat can potentially elucidate genotype-structural-behavioral associations in\nmyriad diseases.\n