STA 290 Seminar Series
DATE: Thursday May 3rd, 4:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm, MSB 4110
SPEAKER: Thomas Nichols, Professor of Neuroimaging Statistics, Oxford Big Data Institute, Nuffield Department of Population Health, University of Oxford, UK
TITLE: “A Journey Into Neuroimaging Statistics”
ABSTRACT: In this informal talk I'll introduce the speciality area that I've worked in for the past 25 years, the development of statistical methodology for human brain imaging. I'll start with a gentle introduction to the brain, how we image it and the sort of questions neuroscientists want to ask with this type of data. Then I'll discuss the type of statistical challenges neuroimaging presents. Some problems that neuroscientists face simply amount to computational ones, scaling up standard statistical models to be fit on 100,000 volume elements (voxels). Others are intrinsically statistical, like modelling neuroimaging meta-analysis data, where only 3D point patterns of locations of changes are available. Finally I'll review what I see as some of the unmet statistical challenges in brain imaging today and the ripe opportunities available for collaboration.
This is a Keynote Lecture ahead of the 2018 Peter G. Hall Conference “Brain and Data Science”, which will take place on Friday May 4th. Please visit http://anson.ucdavis.edu/hallconference2018/ for a full schedule.