Global exploratory analysis of massive neuroimaging collections using Microsoft Silverlight PivotViewer

Microsoft Silverlight PivotViewer is a web service architecture that provides fluid access to massive image databases using digital zoom image technology. The Pivot interface models some aspects of human vision, with the ability both to perform a global scan and to rapidly focus on the finest details. The inclusion of many levels of metadata in an XML file makes it possible for clients to rapidly re-sort and filter images on the web in ways that are powerful and highly informative. We have used Pivot to restructure a very large collection of histological images (2D) taken from the Mouse Brain Library (MBL, www.mbl.org). The new Pivot system (mbl.pivotcollections.org) can be subdivided, filtered, and sorted using a range of continuous and discrete metadata variables (sex, age, strain, genotype, behavior). The current implementation is optimized for visualization, but does not yet have analytic or annotation tiers that would enable facile quantitative dissection of image collections.