Content-based fMRI brain maps retrieval

Abstract

The statistical analysis of functional magnetic resonance imaging (fMRI) is used to extract functional data of cerebral activation during a given experimental task. It allows for assessing changes in cerebral function related to cerebral activities. This methodology has been widely used and a few initiatives aim to develop shared data resources. Searching these data resources for a specific research goal remains a challenging problem. In particular, work is needed to create a global content-based (CB) fMRI retrieval capability. This work presents a CB fMRI retrieval approach based on the brain activation maps extracted using Probabilistic Independent Component Analysis (PICA). We obtained promising results on data from a variety of experiments which highlight the potential of the system as a tool that provides support for finding hidden similarities between brain activation maps.