Department of Biostatistics

People

Prospective Students

Academics

Research

News & Events

Calendar

Consulting

Employment Opportunities

Resource Quicklinks

Computing Environment

Contact

 


ABSTRACT

Template Mixture Models for Image Analysis

Diana Miglioretti, PhD Candidate in Biostatistics

This talk will describe a statistical approach for the analysis of direct cortical electrical interference (DCEI) data with the goal of delineating and comparing cortical regions critical for two language tasks: adult speech discrimination and auditory comprehension. It is hypothesized that speech discrimination is more localized and auditory comprehension more dispersed both within and between individuals.

DCEI is performed with surgically implanted, indwelling, subdural electrode arrays, often covering the lateral left cortical surface. A current passed between a pair of adjacent electrodes creates a temporary lesion during testing, producing a spatial array of binary responses for each patient.

Our statistical approach is to estimate an underlying spatial response function for each patient using a mixture of an unknown number of simple geometric shapes or templates (e.g., circles) with unknown centers and sizes to be estimated. Variation across patients is modeled through random center locations and sizes. The approach will be illustrated with preliminary analysis of the DCEI data and may also have application to other neuroimaging problems.


Return to Longitudinal/Survival Working Group List | Return to Home Page
© 2004, The Johns Hopkins University. All rights reserved.
web policies, 615 N. Wolfe Street, Batimore, MD 21205-2179, 410-955-5000