Mining Lesion-Deficit Associations in a Brain Image Database

Vasileios Megalooikonomou
Department of Computer Science, Dartmouth College

2:00pm Friday, December 10, 1999
Lecture Hall V, ECS

The main goal of the Human Brain Project funded by the National Institutes of Health is functional brain mapping, i.e., discovery of associations between structures and functions of the human brain. For this purpose, large brain image databases that include structural (e.g., magnetic resonance imaging (MRI)), and/or functional (e.g., fMRI) information, and other behavioral data have been developed.

The Brain Image Database (BRAID) was developed at the Johns Hopkins University for the purpose of functional mapping of the human brain through the study of lesioned (abnormal) brain structures and associated deficits. BRAID is a large-scale archive with an analytical query mechanism. It is implemented using the object-relational database scheme and integrates image processing and visualization capabilities with statistical analysis of spatial and clinical data, providing access via extended SQL through a web interface.

In this talk, we present the architecture of BRAID and the mining process for discovering lesion-deficit associations. We discuss issues of scalability and morphological variability. We demonstrate the use of the proposed mining methods by applying them to epidemiological data finding clinically meaningful associations. Furthermore, we evaluate their performance using data produced by a simulator that is based on the bayesian network model.

-------
Vasileios Megalooikonomou is a visiting assistant professor in the Department of Computer Science at Dartmouth College. From 1997 to 1999 he was a Faculty Research Associate at the Neuroimaging Laboratory, Johns Hopkins University. He received the M.S. and Ph.D. degrees in computer science from the University of Maryland, Baltimore County, in 1995 and 1997, respectively. His research interests include multimedia database systems, data compression, and data mining. He is involved in the Brain Image Database Project, part of the Human Brain Project, funded by NIH.