Meaning and Method of Signal Positive Particle Detection using 2D Gaussian

Nature volume528 , pages280–285 (10 December 2015)
https://www.nature.com/articles/nature16443

The part of image analysis in this paper, they detect signal positive particles using Gaussian fitting, which doesn’t rely on intensity threshold. It does not make any sense. Could you describe it, please? And how can I apply this technic to an image?
I attached an example image which has 3 z slices over time.![signal%20positive%20particle%20detection|690x385,50%](upload://tzK7mvsample(1).tif (11.3 MB)
BHFG2aliPbNy4qCA8EARf.jpeg)