Image Informatics for Biological and Biomedical Imaging
Friday, September 14, 2007
11:30am – 1pm
Union South, TITU
Speaker: Kevin W. Eliceiri, Director, Laboratory for Optical and Computational Instrumentation
View Presentation Video
View Presentation Slides (.pdf)
With the increasing use of digital image capture microscopy in the biomedical sciences, it has become a major challenge to locate, view and interpret large numbers of images collected in a diversity of formats. Many biological research laboratories have a pressing need to archive and annotate vast numbers of images collected by video, laser-scanning microscopy and other imaging techniques. Multidimensional images, such as four-dimensional images from multi-focal plane time-lapse recordings, or images from spectral and lifetime microscopy, make the challenge even greater. Without careful organization, important research data can be difficult or impossible to find, much less visualize and analyze effectively.
We are developing a complete, open source system for handling biomedical images, including image acquisition, data storage, metadata (experimental data associated with an image), visualization, analysis, annotation, and database interconnectivity. The software makes extensive use of several established open source systems for handling images: the Open Microscopy Environment (OME; image database),
VisAD (image visualization), and ImageJ (image processing). We have also developed an open file format called OME-TIFF: fully compatible multi-page TIFF with rich metadata embedded in OME-XML format, the OME standard. Other major components of the software include: Bio-Formats,
a Java library for reading and writing dozens of common microscopy file formats, including processing and conversion of metadata into OME-TIFF format; VisBio,
a biological visualization tool for facilitating visualization and analysis of complex multidimensional image data; an OME plugin for ImageJ for transferring images between ImageJ and an OME database; and a multidimensional Data Browser plugin for ImageJ to enable more effective visualization of 5D image data within ImageJ. We are working to integrate these packages into a cohesive whole to provide microscopists with a complete toolset for overcoming the challenges of working with digital multidimensional images.
Any questions may be directed to Hideko Mills or Steve Krogull.