A Conformal and Statistical Mapping Method for 3D Pet Image Analysis

D. Pai, G. Zou, J. Hua, M. Dong, X. Gu, and O. Muzik (USA)

Keywords

Image Analysis, Conformal Mapping, Registration

Abstract

Multimodality brain imaging represents a valuable tool in the study of anatomical and functional characteristics of the brain. Recent advances in imaging technologies, such as Magnetic Resonance Imaging (MRI) and Positron Emis sion Tomography (PET) have significantly accelerated neu roscience research in the last decade and have facilitated an improved understanding of the complex relationship be tween structure and function. This paper presents a novel framework based on global conformal surface mapping and alignment of the brains. The mapping allows accurate and reproducible transformation of each subject’s brain to a canonical spherical domain and is the basis for subse quent statistical analysis of cortical elements (defined on the spherical domain) across patients. The cortical ele ments can be defined at arbitrary resolution, which serve as the units of statistical analysis of PET images and might in future be the basis for quantitative assessment of fiber tract connectivity between anatomical territories of the brain.

Important Links:



Go Back