Within the framework of the interdisciplinary project "Virtual Brain", the computer scientists of JARA-HPC support the neuroscientists of the section JARA-BRAIN in researching the human Connectome. With a specific computing method, the so-called probabilistic tractography, main fiber pathways as well as narrow or crossing fiber bundles can be recognized and uncertainties in the underlying measured data determined. On the basis of the gathered data, JARA-HPC scientists create a 3D visualization of the human brain's fiber pathways, thus enabling the neuroscientists to understand the interactions of different brain areas more profoundly.
Neuroscientific research aims at understanding the structure-function relationships in the brain. Networks of communicating brain areas are required to fulfil motor, sensory and all mental and cognitive activities. The structural basis of such networks are nerve fibers connecting the partcipating brain areas. A profound knowledge about this connectivity structure of the brain is therefore necessary for understanding the computational activity of the brain. Consequently, the concept of the human Connectome has recently evolved, a research strategy comparable to the Human Genome Project. One major goal of the Connectome Project is a complete mapping of the nerve fibers and fiber bundles in the brain. Understanding of the Connectome is also required as the anatomical basis for a sufficient understanding of psychiatric or neurological diseases.
The concept of the human Connectome constitutes a field of research where the knowledge of the connectivity and fiber architecture of the brain is the structural and indispensable basis for understanding functionally interacting brain networks. Due to the integration of real-time rendering and interaction techniques of 3D visualization, the interdisciplinary cooperation of JARA-HPC and JARA-BRAIN enables the control and continuous improvement of the data on which the fiber pathway modeling is based. At the same time, visualization increases the intuitive understanding of the underlying mathematical models and multimodal information.
We expect that the cooperation of JARA-HPC and JARA-BRAIN will provide the opportunity to extend the knowledge of the brain from the macro to the micro level, resulting in a more detailed understanding of the structures and connections within the brain. Extended knowledge in this field could lead to new insight into certain neurological disorders and diseases that can be traced back to interrupted fiber pathway connections.
JARA-HPC Seed Fund Project "Collaborative Exploration and Visualization in ivNet"
In order to provide a distributed analysis of scientific phenomena between researches from Aachen and Jülich, methods for direct remote data access and for collaborative visualization components are required.
This project has a special focus on the concept of persistency for gaining and distributing insights among scientists.
A medical application for the visualization of diffusion tensor imaging (DTI) data, which are used to investigate anatomical connectivity in the living human brain, will serve as major test platform.
The goal of this project is to provide methods for so-called co-temporal collaboration where actions are not only geographically collocated but where asynchronous work contexts are becoming an important issue.
"Interactive Visualization of Uncertainty in Probabilistic Tractography of Brain's White Matter Pathways as assed by Diffusion Tensor Imaging", Anette von Kapri, Tobias Rick, Svenja Caspers, Simon Eickhoff, Karl Zilles, Torsten Kuhlen in "SPIE Medical Imaging", February 2010
"Accelerating Radio Wave Propagation Algorithms by Implementation on Graphics Hardware", Tobias Rick, Torsten Kuhlen in "Wave Propagation" 2010, ISBN 978-953-7619-X-X
"GPU Implementation of 3D Object Selection by Conic Volume Techniques in Virtual Environments", Tobias Rick, Anette von Kapri, Torsten Kuhlen in "IEEE Virtual Reality Conference", March 2010
Dr. Bernd Hentschel
Dr. med. Svenja Caspers
Please download the project's brochure here.
You can watch the project's movie here.