Skip to content

Helper macros for certain image analysis tasks in ImageJ

License

Notifications You must be signed in to change notification settings

jhyot/imagej-helper-macros

Repository files navigation

ImageJ Helper Macros

A set of ImageJ macros I wrote during my PhD thesis to assist with fluorescence intensity measurements. Currently there are 3 different macros, each in its own file. These macros can be imported and run via the usual ImageJ macro commands (refer to the ImageJ manual/wiki for a description of how to run macros).

copy-timings-to-clipboard.ijm

This macro analyses any optical microscopy time series file and copies the acquisition times of each frame to the clipboard. After starting the macro, select the time series file you wish to analyse, and then select which series and channels you’d like to include in the output. After the macro finishes, the timings are copied into the clipboard in a tab-separated format (a column for each selected channel), so that it can be easily pasted e.g. into a spread sheet.

The Bio-Formats plugin for ImageJ must be installed for the macro to work.

measure-ring-intensities.ijm

This macro measures the fluorescence intensity over a regular ring/dot pattern. See (Hyotyla, 2016, pp. 49, 91) for the kind of pattern that this macro was written for.

The macro can handle a single image or a stack of images. Upon running the macro a dialog window will ask for various parameters. The grid angle and position of the top-left ring can optionally be determined in a visual way: Draw exactly one line ROI horizontally across a row of rings (determines the angle), and one point ROI in the center of the top-left ring (determines the x and y starting coordinates).

Parameters can also be read in from a file. The parameter file can be chosen in the next window appearing after closing the parameter dialog. In the parameter file, parameters must be defined one per line in the format parameterName=value. For a list of all parameters look at the header of a result file after running the analysis once. Missing or invalid parameter values will be replaced by a default value.

Fluorescence intensities can be adjusted for background by two different methods (the third option is no background adjustment).

  • median min

    • The median min method runs a median filter over the image to reduce noise outliers and takes the minimum intensity value around each ring as the background for that ring. This background value is subtracted from the ring intensity to arrive at the background-corrected ring intensity.
  • avg normalized

    • The avg normalized method takes the average intensity (no filtering applied) within a background ring region (bounded by an inner and outer radius as defined in the analysis parameters), and normalizes the nanoring intensity by this background average for each ring.

roi-multi-measure.ijm

This is a small helper macro to quickly measure all ROIs for all slices in all open images. The user can define ROIs for a single image, and then automatically have the ROIs measured in all slices and images. For this to work, the ROIs must not be named in a slice-specific way, i.e. the sss-yyyy-xxxx format is not allowed. The macro writes the pixel area, average intensity, standard deviation, and integrated intensity into a single file.

References

Hyotyla, Janne T. Nanomechanics of Confined Polymer Systems. 2016, PhD Thesis, University of Basel, Faculty of Science. doi: 10.5451/unibas-006622415 (download)

License

This software is licensed under the MIT license. See the LICENSE file for full details.

Copyright (c) 2016 University of Basel

Acknowledgements

Macros originally written by Janne Hyötylä, [email protected]

About

Helper macros for certain image analysis tasks in ImageJ

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published