Visualization in Connectomics - Scientific Visualization: Uncertainty, Multifield, Biomedical, and Scalable Visualization - page 219

21.7.1 Brain Mapping
A major goal in neuroscience is to define the cellular architecture of the brain.
Mapping the fine anatomy of complex neuronal circuits is an essential first step
in investigating the neural mechanisms of information processing. The term brain
mapping describes a set of neuroscience techniques predicated on the mapping of
biological quantities or properties onto spatial representations of the brain resulting
in maps. While all of neuroimaging can be considered part of brain mapping, the
termmore specifically refers to the generation of atlases, i.e., databases that combine
imaging data with additional information in order to infer functional information.
Such an undertaking relies on research and development in image acquisition, repre-
sentation, analysis, visualization, and interaction. Intuitive and efficient visualization
is important at all intermediate steps in such projects. Proper visualization tools are
indispensable for quality control (e.g., identification of acquisition artifacts and mis-
classifications), the sharing of generated resources among a network of collaborators,
or the setup and validation of an automated analysis pipeline. Data acquired to study
brain structure captures information on the brain at different scales (e.g., molecular,
cellular, circuitry, system, behavior), with different focus (e.g., anatomy, metabolism,
function), and is multi-modal (text, graphics, 2D and 3D images, audio, video) [ 15 ,
53 ]. The establishment of spatial relationships between initially unrelated images
and information is a fundamental step towards the exploitation of available data [ 7 ].
These relationships provide the basis for the visual representation of a data collection
and the generation of further knowledge.
Databases and atlases. Aneuroanatomical atlas serves as a reference frame for com-
paring and integrating data from different biological experiments. Maye et al. [ 64 ]
give an introduction and survey on the integration and visualization of neural struc-
tures in brain atlases. Such atlases are an invaluable reference in efforts to compile a
comprehensive set of anatomical and functional data, and in formulating hypotheses
on the operation of specific neuronal circuits.
A classical image-based neuroanatomical atlas of Drosophila is the FlyBrain
atlas, 1 spatially relating a collection of 2D drawings, microscopic images, and text.
One approach in generating a digital atlas of this kind is by acquiring confocal
microscope images of a large number of individual brains. In each specimen, one
or more distinct neuronal types are highlighted using appropriate molecular genetic
techniques. Additionally, a general staining is applied to reveal the overall structure
of the brain, providing a reference for non-rigid registration to a standard template.
After registration, the specific neuronal types in each specimen are segmented, anno-
tated, and compiled into a database linked to the physical structure of the brain. Jenett
et al. [ 40 ] describe techniques for quantitative assessment, comparison, and presen-
tation of 3D confocal microscopy images of Drosophila brains and gene expression
patterns within these brains. Pereanu and Hartenstein [ 71 ] and Rybak et al. [ 79 ]
described 3D atlases of the developing Drosophila brain and the honeybee brain.
1 http://flybrain.neurobio.arizona.edu