Data in Summary Table are cleared randomly from the top

Cheers.

A while ago, with the help of the ImageJ community, a macro was developed that linked the ROI manager to the Analyze Particle function, for measuring particle-like structures in an image stack with varying ROIs from slide to slide. Here’s a link for more info:

I have used the following macro for more than 300 samples (about a year ago) without any problem:

numROIs= roiManager("Count");
imgID= getImageID();
for (j=0; j<numROIs; j++)
{
	selectImage(imgID);
	roiManager("select",j);
	run("Duplicate...", " ");
	run("Analyze Particles...", "size=500-infinity circularity=0.0-1.00 show=[Outlines] display include summarize");
	run("Close");
}
selectImage(imgID);
run("Close");

However, running the macro again, I encounter this issue that the data in the summary table are cleared from the top. I have repeated the same procedure, for the same sequence of images, and every time during the run, I see data suddenly disappearing from the top of the summary table, like somebody has just clicked a refresh button!

Here, for example, the data on the left (Summary_1.xlsx) are from a previous measurement when everything was working correctly (done last year). On the right, for the same sample, I have lost the summary data of the first 181 slides.

The strange thing is that the number of missing data seems to be totally random. Upon one run, I lose 40 rows of data, when I run the macro another time (for the same image sequences), I lose the data for 75 slides, and it the above-mentioned example, I lost the data for 181 slides.

Does anybody have an idea what might be wrong? I use ImageJ 1.53c.
Any help would be much appreciated

Do you see the problem when you use the Help>Upgrade ImageJ command to downgrade to ImageJ 1.52?

Hi Wayne,

This morning I tried a couple of older versions. 1.53, 1.52a, 1.52p (fiji), and 1.51m had the same issue.
1.51n (Fiji; 20170530) however, worked without any issue!

Can it be a Java-related problem on my PC?

Could be, or it could be an ImageJ regression. It would help if you could provide a small, but runnable, macro that reproduces the problem.

I tried to reproduce the problem with the following macro, but it works as expected. This bug should be easy to fix as long as there is a way to reproduce it.

  for (i=1; i<=10; i++) {
     run("Blobs (25K)");
     rename("Image "+i);
     setAutoThreshold("Default");
     run("Analyze Particles...", "display exclude clear summarize");
     close;
  }