Joint Sparse Representation of Brain Activity Patterns

Abstract

Neurological disorders that affect brain structure, function and networks would substantially benefit from developing new techniques that combine multi-modal and/or multi-task information. We have proposed a Joint Sparse Representation Analysis method to identify common information across different functional contrasts in data from an fMRI experiment. We evaluated the use of a sparse representation analysis method within a Fisher Linear Discriminant (FLD) classification to classify individuals as young or older, based only on functional activation patterns in a speech listening task. Sixteen young (age: 19-26) and 16 older (age: 57-73) adults were scanned while listening to noise and to sentences degraded with noise, half of which contained meaningful context which is known to enhance intelligibility. Functional contrast images representing different perceptual and cognitive components of speech perception (i.e., auditory perception; speech perception, use of context to enhance perception) were used within the joint sparse representation analysis to generate basis activation sources and their corresponding sparse modulation profiles. Sparse modulation profiles were used to classify individuals into the young and older categories. Results demonstrated that a combination of functional contrast images yielded excellent classification performance.

Papers

---

---

Mahdi Ramezani

Last modified: Nov 16, 2012