System Development Realities:
When the Integrated Microscopy Resource 4D team started this project a few years ago, our goal was to develop an inexpensive, easy-to-use system that could be made available to as many researchers as possible. We also wanted to take advantage of existing technology rather than reinventing it. With these requirements in mind we decided on the Macintosh computer as the platform that best met our needs. There were several tools and technologies available on the Mac that simplified and accelerated development.
Development Tools:
One tool that we found extremely useful was the public domain program NIH Image. The powerful macro language allows you to add features. If you need functions not available through macros, the source code is available and easily modified. There is also a large user base and very active Internet mail list if you have questions. Charles Thomas at the IMR used a combination of macros and source code additions to NIH Image to develop the first versions of the 4D acquisition software.
The most significant technology that aided our development was an operating system extension called QuickTime. This software adds support for compression and dynamic data formats to the Macintosh operating system. This includes more than just the ability to compress images and play movies; it adds very powerful tools for dealing with dynamic, linked multi-track data.
QuickTime's ability to compress and decompress images on the fly was essential to our project. This code was efficient enough to achieve high frame rates on standard inexpensive hardware. QuickTime has support for several different compression algorithms, but we found JPEG compression to give us the best combination of file size, image quality and frame rate. We also found that noise reduction by frame averaging on the DSP 2000 helped increase compression ratios.
We would normally expect that a complete C. elegans development series would consist of as many as 18,000 frames. Uncompressed, this data would consume 6 Gigabytes of disk space. Datasets of this size would not only consume large amounts of expensive online storage, it would also make it extremely difficult to move the data across a network or distribute to other researchers. By using QuickTime's JPEG compressor we are able to achieve compression ratios of 15:1 while still retaining good image quality.
In order for users to get the most out of this data, I developed the initial viewer application called 4D Navigator. The 4D Navigator lets the users move through these datasets, set bookmarks and go to specific frames. Users choose a data set using the menu, and can move both forward in time and backward in time, move up and down in focal plane and even diagonally. The program allows users to move between frames at a high rate despite the fact that the JPEG compressed images are being read off of the hard drive and decoded on the fly. As users move from one image to the next the frame and focal plane coordinates are updated in the windows at the bottom of the screen. Users can select a specific frame, and set four different bookmarks.
Engineers at Apple Computer used the 4D Navigator at several universities to demonstrate how the Macintosh can be used for multimedia. They also asked if the 4D Navigator program and example data set could be put on an Apple demonstration CD-ROM for the Medical and Science markets.
ftp2.bocklabs.wisc.edu
login: anonymous
password: youremailaddress
cd pub/lmb/imr/4D_Viewer/Viewer/
Charles has also created an excellent overview of the 4-D system which can be found at the following URL:
http://www.bocklabs.wisc.edu/imr/facility/4D/4d.htm
Thomas C., DeVries P., Hardin J. and White J. (1996) Four-Dimensional Imaging: Computer Visualization of 3D Movements in Living Specimens Science 2 August 1996 273, 603-607