fMRI-DTI modeling for prediction and detection of fiber tracts in patients with mass lesions: initial results
Speaker: Lauren O , Harvard Medical School
Date: February 11 2010
Time: 3:00PM to 4:00PM
Location: 32-D507
Host: Polina Golland, CSAIL
Contact: Polina Golland, x38005, polina@csail.mit.edu
Relevant URL: The goal of this research is to design statistical atlas models that
can be created from normal subjects, but can generalize to be
applicable to abnormal brains. Individualization of atlas information
is commonly addressed using sophisticated image registration and
segmentation. However, these methods may not generalize well to
neurosurgical patients with mass lesions. In contrast, we propose to
develop a robust atlas that captures relationships that are relatively
preserved across subjects, even in the case of mass lesions. In this
work we present a rotation and translation invariant model that
defines the spatial relationship between fiber tracts and
functional/structural landmarks. We show that the model can be used to
predict and detect fiber tracts and we successfully apply it to a
small group of patients with mass lesions.
See other events that are part of Biomedical Imaging and Analysis 2009/2010
See other events happening in February 2010
