TrakEM2 software for neural circuit reconstruction:

Start from a raw data set of images of serial sections, imaged with electron microscopy (EM) or laser-scanning microscopy (LSM). Then automatically montage and register all image tiles within a section and across serial sections. Then apply manual and semiautomated methods for segmenting all objects of interest, be it cell bodies, neuronal arbors, synapses, blood vessels, neuronal lineages, nerves, glial cells, or whatever objects recognizable by a human. All segmentations are organized in hierarchically nested, meaningful groups; for example, a collection of branches and synapses are collapsed under a "neuron" node, and a group of neurons are collapsed under a "lineage" node, etc., where nodes are arbitrarily defined as demanded by the analysis. Means are provided to visualize reconstructions and image volumes in 2d/3d, to count and measure them, to analyze them using custom-built scripts, and to export them in a variety of formats including CATMAID (for visualization via the web), Wavefront and NeuroML.