Voxlab is a visualization tool for multi-modal inter-subject neuroimaging.
The software is currently in internal test-use phase. Please check back to this page later for more information.
Gallery
Here you can find some sample images from Voxlab:
Aim
Modern brain imaging methods (e.g. MEG, MRI, fMRI, DTI) produce a
wealth of information across subjects and imaging modalities. Novel
analysis methods are constantly under development and they, as well as
the already established analysis methods, employ various software,
file formats, and coordinate systems. The real challenge is to bring
all this heterogeneous data together for joint visualization to
facilitate visual inspection and conclusions.
Voxlab is designed to:
- be able to read and use various data types (e.g. volume data,
triangle meshes)
- place minimal requirements for the visualized data (e.g. to allow
inhomogenious voxel sizes and arbitrary number of voxels along each
axis)
- employ top-quality processing: minimum use of conversion and
resampling steps ⇒ minimal processing and resampling artifacts
(e.g. data kept and visualized in the original format, such as
floating-point values, and in the original coordinate system)
- support natively (through plugins) all the file-formats used in
the field ⇒ no file format
conversions needed ⇒ no unnecessary copies of data and less steps
to err
- use plugin architecture which allows extension of capabilities
by others
- employ strong identification and robust matching of subjects,
coordinate systems, and various coordinate transformations in the data
⇒ easy and safe operation for the user
- have user-friendly graphical user interface (GUI) with
e.g. drag-and-drop functionality to allow intuitive real-time manipulation of
data
- use industry-standard OpenGL graphics with GLSL (OpenGL shading
language) for visualizations ⇒ GLSL code generated and compiled
on the fly ⇒ the power of modern GPUs is well utilized
- run on Linux and Mac OSX platforms with multi-threading to benefit
from multi-core CPUs
Advanced Visualization Options
- arbitrary number (up to hardware maximum, typically 6) of
orthogonal or non-orthogonal clip-planes, both for volume and mesh
rendering
- full-scene anti-aliasing for high-quality output
- camera depth-of-field (DOF) effect for better control of viewer's
attention and focus in complex scenes
- mesh carving: a surface can be carved out from another surface at
visualization time (e.g. skull's outer surface can be carved hollow
with the inner skull surface)
- playback: time-dependent data and stimuli can be played back at
selected speed, possibly restricted to chosen intervals only
- support for 3D displays (implemented, but not yet thoroughly
tested and enabled)
