/AutoScanJ

A suite of ImageJ and Micro-Manager scripts to perform intelligent imaging

Primary LanguageJavaApache License 2.0Apache-2.0

AutoScanJ

A suite of ImageJ and Micro-Manager scripts to perform intelligent imaging.

AutoScanJ is a suite of ImageJ scripts enabling to image targets of interest by automatically driving a motorized microscope (controlled by Micro-Manager or Leica LASAF with Matrix/CAM modules) at the corresponding locations. For live samples, the software can detect biological events from their onset and sequentially image them at high resolution, an action that would be impractical by user operation. For fixed samples, the software can dramatically reduce the amount of data acquired in situations where statistically few objects of interest are observed per field of view.

The target detection function can be simply customized in ImageJ macro language and AutoScanJ can run directly on the microscope workstation, or on a dedicated image processing workstation connected to the same local network as the microscope workstation. In this latter case, the interaction is implemented as network socket-based messages and the images are shared by network folder (instead of a local folder).

The concept of intelligent microscopy and sample applications corresponding to the different possible scenarios (fixed/live sample and tiled/block scans) is described in detail in the article AutoScanJ: A Suite of ImageJ Scripts for Intelligent Microscopy.

The code has only be thoroughly tested with Fiji Life-Line 02 June 2014 and Windows operating systems. The code might run with other configurations but there is no guarantee (missing plugins, file path separators). Also, as a general run when using ImageJ, you should refrain from using space separators in file and folder names.

AutoScanJ Software and Hardware Components: https://bit.ly/3nSfFAA
Manual to write custom target detection functions: https://bit.ly/2LJzLPN
Software documentation for use with Micro-Manager: https://bit.ly/2SyWcHS
Software documentation for use with Leica LAS AF: https://bit.ly/2GEhu4a

Don't have a compatible microscope at hand but still want to test the macros?

Download test data here: https://bit.ly/3d25TYt

Usage of the test data

For Fixed + Tiling experiments:

  • Unzip the data that you want to test to an empty folder
  • Run the corresponding ImageJ macro and select the folder you unzipped the data to as experiment folder
  • Untick "Perform primary scan" and "Send CAM scripts" so that the macro does not attempt to re-acquire primary scan image by controlling the microscope
  • Select the Automatic pre-analysis function corresponding to the dataset you downloaded (Glomerulus_detector or Metaphase_detector)
  • Leave all other options to there default

You should see the primary scan map with the detected targets and the targets should be montaged in the QuickView window. It is not possible to go further with this demo and acquire secondary scan images since the microscope is not connected. Pressing OK twice in the next dialog boxes will bring the expected message "The images from the CAM list cannot be found" since no secondary scan images will be acquired. In a regular experiment with microscope connected, the targets should exclusively be edited from the Quick View window (if enabled in the initial dialog box), or from the primary scan map (if Quick View is not enabled).

For Fixed + Blocks experiments:

  • Follow the previous instruction but select the Automatic pre-analysis function Cytoo_Isolated_Nuclei_SP5
  • At the first dialog box draw a rectangular region inside the Cytoo block (that is around the inner 144 patterns with some black margin around)
  • At the next dialog box approximately circle a pattern
  • Press Ok at the dialog box "Debug mode: isolated cells"
  • The isolated cells of the current block are reported as yellow crosses
  • The next three blocks can be analyzed and inspected likewise by each time pressing OK in the next dialog boxes

For Live + Tiling experiments:

  • Unzip the data that you want to test to an empty folder
  • Create another empty folder (required to empty it before each run)
  • Set the variable OfflineFilesPath in the corresponding IJ macro to the path of the folder you unzipped the data to
  • Run the ImageJ macro and select the second empty folder as experiment folder
  • Untick "Send CAM scripts" so that the macro does not attempt to re-acquire primary scan image by controlling the microscope
  • Leave repetition period to default or increase it to larger values (e.g. 15 or 30 seconds if your computer is rather slow)
  • Leave the Automatic pre-analysis to the default Mitosis_Microtubulin

Images from the primary scan will be pulled peridoically from the first folder and copied to the experiment folder to mimick primary scan acquisition. Mitosis detections should be detected after few frames (as yellow circles overlaid on top of the primary map). No secondary images will be acquired and displayed since there is no connection to the microscope.

For Live + Blocks experiments:

No sample data available yet.