VII Center for Visual Informatics and Intelligence wsu

Collaborative Research: Integrated Modeling and Learning of Multimodality Data across Subjects for Brain Disorder Study



With ever-improving imaging technologies and ever-increasing high-performance computational power, the complexity and scale of acquired brain imaging data have continued to grow at an explosive pace. Rapid advances in multimodality imaging technologies have significantly accelerated brain disorder studies by providing complementary information on many aspects of the human brain in the normal and diseased states. Capitalizing on the availability of large-scale data, we are now able to computationally integrate, index and model the brain functions across a large population for discovering more detailed understanding and more profound knowledge about complex biological interactions in the human brain. Based on our continuous research effort along this direction, this NSF project is developing a novel, rigorous theoretical framework based on Riemannian geometry, multivariate simplex splines, and statistical learning, which provides a basis for multimodality information integration and understanding across populations. Specifically, our research team will design a fundamental framework for advanced and integrated analysis of brain imaging data. It is expected that the developed, advanced informatics tools will allow the quantitative and integrative analysis of a variety of functional patterns and the relationships between anatomical and functional features in different datasets. The proposed computational framework has the potential to be applied across multiple areas of brain research as well as in clinical diagnosis.

The above figure shows the conformal brain surface model (Figure A and Figure B) facilitates accurate matching and registration among subjects in the canonical, spherical domain, hence supporting integrated cross-subject analysis of Positron Emission Tomography (PET) (molecular-level brain activity analysis) (Figure C), Diffusion Tensor Imaging (DTI) (neural fiber connectivity analysis) (Figure D), and Electroencephalography (EEG) (time-varying signal analysis) (Figure E) in computer-aided diagnosis of brain disorders.


Further details of this work are available soon.

Contact Webmaster
Center for Visual Informatics and Intelligence(VII) © 2009