Fish gonad histology analysis

[Kelsie Miller]

Hi everyone,

I am currently struggling to do some image analysis of juvenile fish gonads
that I work with. My goal is to be able to distinguish between male tissue
(spermatocytes, small black dots), female tissue
(oocytes, dark purple with light pink inside), and
undifferentiated tissue (light pink tissue surrounding some of the oocytes
and spermatocytes) within intersex fish (intersex fish image). My hope is
to be able to say the percentage of male, female and undifferentiated
tissue within each gonad.

I am new to doing image analysis but have played around with my images
quite a bit and asked around my university for help to no avail. The main
problem is that I am unable to create contrast between the dark purple of
the oocyte and the dark black of the sperm in order for ImageJ to select an
area by color (see intersex fish image). I do not think I will be able to
have ImageJ analyze by size or shape due to some small gaps within the
gonads and the variety of shapes and sizes of the oocytes. I have had
success at getting the overall area of the gonad, just not the subsections.

My back-up plan is to hand-select all of the different tissue types
although I am hesitate because I feel like this is prone to human error and
extremely labor intensive.

Any suggestions are greatly appreciated!

Thank you,
Kelsie

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

[Straub, Volko A. (Dr.)]

Dear Kelsie,

Have you tried Gabriel Landini's colour thresholder available in
Fiji-ImageJ, which might be useful for your attempted colour separation?
It would also be useful to provide an annotated image to make it clear
what you are trying to analyse (perhaps of a subsection of your image) -
usually much clearer than trying to describe what you try to achieve.

Volko

On 08/05/2017 19:21, Kelsie Miller wrote:

> Hi everyone,
>
> I am currently struggling to do some image analysis of juvenile fish gonads
> that I work with. My goal is to be able to distinguish between male tissue
> (spermatocytes, small black dots), female tissue
> (oocytes, dark purple with light pink inside), and
> undifferentiated tissue (light pink tissue surrounding some of the oocytes
> and spermatocytes) within intersex fish (intersex fish image). My hope is
> to be able to say the percentage of male, female and undifferentiated
> tissue within each gonad.
>
> I am new to doing image analysis but have played around with my images
> quite a bit and asked around my university for help to no avail. The main
> problem is that I am unable to create contrast between the dark purple of
> the oocyte and the dark black of the sperm in order for ImageJ to select an
> area by color (see intersex fish image). I do not think I will be able to
> have ImageJ analyze by size or shape due to some small gaps within the
> gonads and the variety of shapes and sizes of the oocytes. I have had
> success at getting the overall area of the gonad, just not the subsections.
>
> My back-up plan is to hand-select all of the different tissue types
> although I am hesitate because I feel like this is prone to human error and
> extremely labor intensive.
>
> Any suggestions are greatly appreciated!
>
> Thank you,
> Kelsie
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
>
> ---
> This email has been checked for viruses by AVG.
> http://www.avg.com

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