Methodological platforms

The Method Platforms; Imaging and Computational Modelling, constitute the foundation for much of the research performed at SBI.

KI was a pioneer in brain imaging in the 80’ies and has several independent groups that provide internationally competitive research. Expertise exists today in MR (magnetic resonance) for functional neuroimaging, diffusion imaging, structural high resolution imaging, parallel imaging, perfusion imaging and image reconstruc tion/post processing. Indeed, human psychology and cognitive functions have become part of brain science, since one can now identify which areas are activated during different cognitive tasks.

Molecular imaging with PET (positron emission tomography) has a proud history at KI, which has contributed more new radioligands suitable for in vivo imaging than any other laboratory in the world. Recent methodological advancements include “best in class” quanti fication and image analyses of minute protein concentrations (subpicomolar) throughout the brain including the neocortex. Applied work has shown that the in vivo biomarkers for the dopamine and serotonin systems have an established potential in research on drug targets, fundamental functions such as reward systems, personality psychology, as well as in research on major psychiatric disorders. Recently a HRRT PET scanner was installed at the PET centre. In conjunction with these methods, high resolution EEG is presently incorporated in the methods of the MR centre.

The target for our computational modelling is to connect data from behaviour and macro scopic measurements (functional imaging) with neuronal, synaptic and receptor level pro cesses by means of a suite of computational models. Targeting such large-scale phenomena, e.g. global cortical dynamics, makes it necessary to use high-end parallel simulation tools running on parallel hardware, like the Blue Gene at SBI. To complement biophysically detailed simulations (SPLIT type) we will perform more abstract ones, or mix detailed and abstract model components in the same simulation.