Login  Register

Re: Nuclei segmentation

Posted by gankaku on Mar 17, 2014; 6:56pm
URL: http://imagej.273.s1.nabble.com/Nuclei-segmentation-tp5006953p5006958.html

Hi Chin, Johannes, William and Adam,

In such a case the watershed irregular features will unfortunately hit its
limit. It does the job as long as the connections between features are
"thin" compared to the features themselfes and to further protrusions. This
is unfortunately not the case for those nuclei.

Besides the fact, that I need to improve its performance I was also
thinking already about additional methods which might lead to a seperation
of features where the normal watershed fails. So far, no good method came
to my mind to achieve this. So, I am happy about any suggestions and would
try to include them in the toolbox.

cheers,
Jan


2014-03-17 16:56 GMT+01:00 Johannes Schindelin <[hidden email]>:

> Hi Chin,
>
> On Mon, 17 Mar 2014, chin wrote:
>
> > Are there better ways to segment those various types of nuclei in a
> image?
> > I tried some threshold methods to segment then use watershed to split
> > connect nuclei. But it seems not be segmented completely by using those
> > methods.
>
> You might need to apply some background subtraction with a real large
> diameter for the threshold, and then use the "Watershed Irregular
> Structures" command in the BioVoxxel Toolbox:
>
> http://fiji.sc/BioVoxxel_Toolbox#Watershed_Irregular_Structures
>
> Ciao,
> Johannes
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>



--

CEO: Dr. rer. nat. Jan Brocher
phone:  +49 (0)6234 917 03 39
mobile: +49 (0)176 705 746 81
e-mail: [hidden email]
info: [hidden email]
inquiries: [hidden email]
web: www.biovoxxel.de

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html