How do I measure oocytes with the nucleus on a histological section while excluding those without the nucleus?

Hi,

I am struggling to find a way to measure oocytes on histological sections with the nucleus while excluding those without the nucleus. Since the measures would involve the whole histological section (i.e. a few hundreds of oocytes to manually be measured) I am trying segmenting and/or thresholding the pictures on imageJ, but without success. The idea underlying this approach is to find a way to automatically detect the “objects” I am specifically interested in.

I’ve already tried ImageJ Trainable Weka Segmentation and MorphoLibJ.

Here a couple of pictures of ovaries DAPi-stained and what I’d roughly like to obtain. Red and yellow circles for oocytes with and without nucleus, respectively. I also upload here another picture of histological section stained with Hematoxylin-Eosin to give you a better idea of the purpose.

Do you have any ideas?

Thank you in advance!