Visual mining of neuro-metaspaces

Large scale neuroimaging data archival protocols are gradually becoming ubiquitous in both research as well as clinical settings. Current user-database interfaces are limited to textual searches and often require data-specific knowledge for performing queries. This is proving to be an obstacle for researchers who wish to obtain a holistic view of the data before designing pilot neuroscientific studies or even formulating statistical hypotheses. Instead of providing a restricted, unidimensional view of the data, we seek to place a multi-dimensional view of the entire neurodatabase at the user's disposal. With the aim of visual navigation of complete neuro-repositories, we introduce the concept of brain meta-spaces. The meta-space models the implicit nonlinear manifold where the neurological data resides, and encodes pair-wise dissimilarities between all individuals in a population. Additionally, the novelty in our approach lies in the user ability to simultaneously view and interact with many brains at once but doing so in a vast meta-space that encodes (dis)similarity in morphometry.