Statistics Seminar: STA 290
Thursday, January 26th, 2012 at 4.10pm, MSB 1147 (Colloquium Room)
Refreshments: 3.30pm, MSB 4110 (Statistics Lounge)
Speaker: Owen Carmichael (Neurology / Computer Science, UC Davis)
Title: Statistical challenges in diffusion MRI and resting state functional MRI
Abstract: Over the past three decades, a great deal of work has gone into understanding the statistical properties of algorithms for processing traditional magnetic resonance images (MRI) of the human brain, which are represented by a regular 3D lattice of univariate measurements. Now, however, neuroscientists are confronted by two novel forms of imaging data: diffusion MRI, represented by a 3D lattice of probability distributions defined on the 3D sphere; and functional MRI, represented by a 3D lattice of time series measurements. Many important
operations on these modalities, including smoothing, alignment, segmentation,
and calculation of population averages are not well understood from a
statistical point of view, and they have a reputation among neuroscientists for
being noisy and unreliable. In this talk I will give an overview of my
collaborations with statisticians and computer scientists who are improving
methods for quantifying time series similarity and producing reliable brain
segmentations from fMRI; and smoothing and tracing neuron paths through
diffusion MRI.
Joint work with Lyudmila Sakhanenko, Hans Mueller, Jane-Ling Wang, Ian Davidson, Jie Peng, and Debashis Paul.