University of Minnesota
IGERT-Neuroengineering Training Program
http://igert-ne.umn.edu/
igert-ne@umn.edu

IGERT-NE home page.

Decoding Brain Signals

Decoding the spatial and temporal complexity of brain activity is a formidable challenge for the field of neuroscience.

While there certainly have been tremendous advances in technologies that can stream large quantities of information from the brain at every scale - from ion channel dynamics using patch clamp recording methods to neuronal spike activity using multi-channel electrode arrays to network states using electroencephalography (EEG) and magnetoencephalography (MEG) to dynamic patterning using functional magnetic resonance imaging (fMRI) - advances in the decoding of brain signals and integrating meaning behind those signals at all levels have not kept pace with the sophistication of data extraction techniques.

Innovation in brain decoding techniques is of great significance for better understanding brain function and if we are to augment or supplement human capabilities, especially when the measurements are of internal, and otherwise inaccessible, cognitive states.

Fundamental research is needed to develop innovative approaches to brain decoding that can extract and integrate large quantities of information embedded in brain signals across multiple spatial and temporal scales.

This IGERT research theme is aimed at advancing state-of-the-art brain decoding, including:

  1. Multi-scale modeling and mapping techniques for understanding how measured signals emerge from dynamical activity in the brain, and

  2. Integrative decoding algorithms that can assimilate multi-dimensional brain recordings across both space and time, and can associate neural activity with quantifiable measures of perception, behavior, and cognition.

 

 

 

 

 

 

 

 

 

 

Left: High-resolution functional mapping of ocular dominance layers in the cat lateral geniculate body at 7 Tesla [Zhang et al, 2010]; Right: Functional connectivity patterns estimated from EEG and fMRI during movement [Babiloni et al, 2005].

IGERT fellows will have opportunities to develop and evaluate novel decoding algorithms, and then use the knowledge gained from these experiments to design and implement the next generation of multi-modal neural interface systems that can be widely used in research and clinical settings. Students will be trained in the fundamentals of brain decoding and neural interfacing approaches through a new core course entitled Neural Decoding and Interfacing, which will guide them through the development of neural decoding systems using theoretical exercises and practical experience.