Science Downloads

plugins, code, resources, etc

Image Processing

MultiStackReg v1.45 - Now just drag/drop into your plugins folder! - One of the simplest and most reliable image alignment methods available in ImageJ is Stackreg. But, in situations with a number of image channels, most of which only detract from the alignment process, it becomes valuable to be able to save and load the alignment parameters. One can then align with the best channel(s), possibly with greatly increased contrast to make the alignment more stable, and apply that transformation directly onto the raw data.

MultiStackReg is principally an alteration to Stackreg to allow for this save/load functionality, but it also enables stack registration. In circumstances where specimens undergo multiple imaging sessions, it is necessary to "register" the stacks (as opposed to aligning), such that each z-slice can be overlayed with the corresponding slice of the other stack. This is done by setting the reference stack in the first position with "use as reference," and then aligning the second stack to it.

Coding note: the file_1 and file_2 fields are there principally to allow for easier usage in a macro. Setting them to a filename with "save" or "load" actions will cause the plugin to save or load from that file, respectively. It's a little clunky, due to the way ImageJ's parameters work, but it works well enough in practice.

Array Tomography

Array Tomography Toolbar v. 09.12.14 for Axiovision - Array Tomography ribbons have a rather interesting layout that defies most imaging applications. AT sections are laid consecutively on a slide in a mostly predictable fashion, but include too much variation for pure extrapolation to track them. We've developed a method for tracking ribbons using short extrapolations refined by cross-correlation searches. When run, this package will add a toolbar to your Axiovision installation (you may have to enable it) containing two versions of our tracking algorithm. To run, unzip and run setup.exe.

The quickest way to get started is to position your camera above a point of interest, press "firstPoint," then move the camera above the corresponding point of the next section and press "secondPoint." The process should be entirely automated from that point, and when it stops you should have a .csv file on your desktop called "Array.csv" You can import this as a position list (see the Axiovision documentation for "Mark and Find") and use it from there. Once you've imaged a position list this way, call "makeStack" to turn the resulting .zva folder into a single volume image.

Synapse Classification - The culmination of my graduate thesis was the use of interactive machine learning to classify synapses in array tomographic volumes.