Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Generative embedding for model-based classification of fMRI data

Linear and nonlinear models for classification have been increasingly used to predict brain states from measures of brain activity obtained through functional magnetic resonance imaging (fMRI). The practicality of current classifiers, however, is restricted by two major challenges. First, due to the high data dimensionality, algorithms struggle to separate informative from uninformative features, resulting in poor generalization performance. Second, most off-the-shelf methods rarely afford mechanistic interpretability. In this talk, I will present a generative-embedding approach that incorporates neurobiologically interpretable generative models into discriminative classifiers. Based on an fMRI dataset acquired from aphasic patients and healthy controls, our approach enables more accurate classification and deeper mechanistic insights about disease processes than conventional methods.
Type of Seminar:
IfA BISON Seminar
Kay H. Brodersen
Department of Computer Science, ETH Zurich
May 17, 2011   13:30

Contact Person:

Elias August
No downloadable files available.
Biographical Sketch: