Can't record macro step. What alternatives?

histology
macrorecorder
plugin
macro
Tags: #<Tag:0x00007fb87e4bebd0> #<Tag:0x00007fb87e4be720> #<Tag:0x00007fb87e4be4a0> #<Tag:0x00007fb87e4be2e8>

#1

I am trying to record a macro to automate a task of a plugin.

The plugin is the IHC Toolbox. When I record a macro I get:

run(“IHC Toolbox”);

Then the Toolbox opens up, but the selections I make in it aren’t recorded.

What am I doing wrong? If the macro record won’t work, what else can I try?


#2

HI @b3nb0z

So… I am not sure that the IHC Toolbox is macro-compatible. According to the IHC Toolbox website - there is nothing documented for macros…

What is the functionality that you are using? Just to see if there are other options in ImageJ that could be macro-compatible for you.

eta


#3

Indeed, it is not. It is meant to be an entirely interactive tool.

@b3nb0z this has been discussed in a couple of other topics already:

As @etarena suggested, there might be other functionality in ImageJ, e.g. Colour Deconvolution or Color Threshold, that might serve your task much better and in a scriptable way. So if you share your goals, we can help you further.


#4

Thank you for the replies!

I am trying to extract the DAB signaling strength from a large number of tissue micro-array images (hemotoxylin-DAB stained). I want to exclude all those that don’t show any, or very low DAB signal.

I have tried the color deconvolution and another plugin, the IHC profiler, but none of them worked very well. The IHC toolbox worked perfectly and is the reason I am trying to use it.

The color deconvolution I used was the built in ImageJ, with vector set to H-DAB. Is there any preprocessing I could try to improve it? Read something about that the white balance has to be right, but not sure what to try…

Do you have any other suggestions I could try?


#5

As stated in the documentation of the Colour Deconvolution plugin, you have to do background subtraction with color correction and you should determine your own color vectors in order to get robust performance.

The color correction requirement actually holds true for any method if you want to do batch processing and not handle every image interactively one-by-one.


You should also try some of the open source tools specifically targeting your use case, the quantification of histochemistry images: