About
From Endrov and VWB
Contents |
Endrov: Goals
When we started with spatio-temporal image analysis we faced the following problems:
- No standardized image format. All vendors have their own formats. The lab data organization was chaotic
- The standard open source image analysis program, ImageJ, does not really support 5D
- ImageJ does not support metadata
- ...and the plugin architecture is a mess. Batch processing is difficult. Views cannot easily be extended.
Endrov is written from scratch to solve these issues.
Another problem were the file formats. Our data is too complicated for most formats, and none of the more advanced formats was open. We developed the OST format which is now a tree based object file format. It can store any type of data but is optimized for images. Reading and writing single image panes is fast (independent of imageset size) and there is no need to read everything into memory. We could then handle our very large imagesets.
Endrov: Features
Endrov itself is a plugin architecture, not necessarily just for imaging. This makes it flexible enough to accommodate future needs of extensibility, and we consider the main feature the simplicity of adding new features. The basic things included are:
- 2D and 3D visualization of image data and annotation
- Visualization of 3D isosurfaces mixed with voxels and selected 2D planes
- Support for large data sets (50GB+)
- Detailed support for data compression, lossy and lossless
- Infinite number of channels, 3D data over time
- Batch processing
- Integration with external tools such as Matlab and OMERO
- Image filters and analysis tools
- "Live" (non-destructive real-time) application of image operations
- 5D ROIs (selected regions)
- Modular design, adding new types of data is simple
- Ability to make "fancy graphics" ie movies and renderings
See the users guide for the most recent list of features
VWB: Goals
VWB is a database for the geometry and gene expression of C.elegans. We found a need for a good framework to work with spatio-temporal imagesets. Since these are 5-dimensional (XYZ, time, channel), pretty much no existing software could do it. This in turn led to the Endrov project.
Specifically for VWB, we are aiming for
- Making a precise 3D map of C.elegans at all stages of embryogenesis, and larva if possible
- Mapping neural connections and hopefully also membranes
- Build a database for the spatio-temporal behaviour of genes
- Present all this in a neat way and support data mining
We would like to see other people contributing with spatio-temporal data. If you are interested then contact the TBU group to obtain instructions on how to perform recordings.
VWB: Features
- We have figured out how to improve normal brightfield/DIC microscopes using software to gain in sensitivity and perform advanced automated recordings. Confocals also benefit from the same technique.
- There are plugins to store, edit and visualize cell coordinates, lineage and expression level
- Work is on-going on all forms of automated analysis and querying
Collaboration
We are always interested in new applications of our software. If you
- have some software/hardware that could be integrated/interfaced
- need a rather large set of new functionality for a research project
- have new ideas in general
then contact johan.REMOVE.henriksson@@@ki.se
About Us
The work is conducted by the Thomas Bürglin group at Karolinska Institute, Department of Biosciences and Nutrition.
This website and the software is maintained and developed by Johan Henriksson, johan.REMOVE.henriksson@@@ki.se
