Meeting Banner
Abstract #2687

Cuvature-Based Biomarkers for Dyslexia: T1-Image Based Surface Analysis Shows Statistically Separable Dyslexic Features

Rudolph Pienaar1, 2, Kiho Im3, 4, Mathieu Dehaes3, 4, Sara Smith5, Barbara Peysakhovic5, Bryce Becker5, Nora M. Raschle5, Patricia Ellen Grant, 26, Nadine Gaab, 45

1Radiology, Children's Hospital Boston, Boston, MA, United States; 2Radiology, Harvard Medical School, Boston, MA, United States; 3Newborn Medicine, Children's Hospital Boston, Boston, MA, United States; 4Pediatrics, Harvard Medical School, Boston, MA, United States; 5Developmental Medicine, Children's Hospital Boston, Boston, MA, United States; 6Radiology, Boston Children's Hospital, Boston, MA, United States

We present a curvature-histogram analysis as robust biomarker for characterizing development dyslexia from typically developing subjects.

Abstract PDF

Keywords

adult adults ages automatic automatically band basis beyond blue body border centroid centroids characteristics characterize characterized child children class classes clean cloud clouds cluster clustering collapsed collected combinations comparative conserved considered considers consistent cortical cumulative curvature curvatures cyan defined denote derived detailed development developmental deviation differently distribution dyslexia dyslexic early emergent entire essentially examined expected extrinsic fact family features field finding five frontal full function functions generated generation geometric grant gray green hemisphere hemispheres highly histogram history hospital improved indicated initial insight inspection intersection intervention lobar lobes measures medical medicine mesh moreover morphological move negative note occipital occurrence outcome overlap paint parietal percentage physiological plot populations positive principal properties providing quantified quickly radiology reader reduced robust scanner school seem separable separation sets several shapes side signed significance simple slices smith space statistical statistically strip subject subjects surface surfaces table taken temp temporal together troughs typical understanding vertex view visual whole yellow