point selection and counting procedure

[Thorsen, Anders]

Hi,

We are counting oocytes from whole mount pictures of fish ovaries. We are
counting a lot of samples so our counting procedure should be as efficient
as possible. The majority of the oocytes (the dark vitellogenic ones) are
counted easily using the analyze particle function. However, after the
particle analysis has been run there is still some oocytes that have not
been counted. Mainly this is the previtellogenic oocytes that are
transparent and difficult to threshold on. These oocytes have to be counted
manually on the picture. I find using the point selection tool is in many
cases efficient, however we should only count oocytes that are larger than
150 µm. That means that oocytes around this threshold value has to be
measured using the line tool to decide wether it should be counted or not.
Since this is a new type of measurement ImageJ then want to empty the
results window to make room for the new type of data, and this makes it very
cumbersome. We could of course do all the manual counting using the line
tool, but that would take much more time and we have a lot of samples.

I can think of two solutions to this: 1. Make ImageJ accept several types of
measurements in the Results window at the same time. 2. Let the point
selection tool (the cross) have a selectable size so that it can function as
a size reference when doing the point selection. I have looked for plugins
or macros that could help us with this but haven't found any. Is anybody
aware of something that could solve this problem? Is there somthing I am
doing wrong? Could a plugin be made to solve this?

Background:
We are an ICES working group that is responsible for making egg-production
estimates for mackerel and horse-mackerel in the North Sea and Norwegian Sea
from mature pre-spawning ovaries. The egg-production estimates are then
compared to the amount of pelagic eggs of these species that is actually
found (using plankton net) during the spawning season. From this the total
spawner biomass is calculated, and this again makes an important basis for
deciding the fishing quotas for these species.

thank you

anders

..................................
Anders Thorsen
Dr. Scient.
Senior Scientist
Recruitment Biology and Behaviour
Institute of Marine Research
Nordnesgaten 50
P. O. Box 1870 Nordnes
N-5817 Bergen, Norway
Phone: [phone]
Fax: [phone]
E-mail: [hidden email]

[Volker Baecker]

Hi,
just an idea: Can't you use the Roi Manager to save your point selection?
You select points. Now you want to measure a distance, so you add your
point selection to the roi manager. You make the line selection and
write the
distance to the results table. Then you restore your point roi from the
roi manager and you continue to add points (using shift).
When you finished, you measure the point roi and get your count as the
number of rows in the results table.
You could as well put line rois of a specific size into the roi manager
and use them as reference.

This is more a workaround then a real solution.
Maybe it would be possible to adapt the new version of the cell counter
plugin
http://rsb.info.nih.gov/ij/plugins/cell-counter.html
to allow for length measurements during the process of counting.

Hope this helps,
Volker
Thorsen, Anders a écrit :

> Hi,
>
> We are counting oocytes from whole mount pictures of fish ovaries. We are
> counting a lot of samples so our counting procedure should be as efficient
> as possible. The majority of the oocytes (the dark vitellogenic ones) are
> counted easily using the analyze particle function. However, after the
> particle analysis has been run there is still some oocytes that have not
> been counted. Mainly this is the previtellogenic oocytes that are
> transparent and difficult to threshold on. These oocytes have to be counted
> manually on the picture. I find using the point selection tool is in many
> cases efficient, however we should only count oocytes that are larger than
> 150 µm. That means that oocytes around this threshold value has to be
> measured using the line tool to decide wether it should be counted or not.
> Since this is a new type of measurement ImageJ then want to empty the
> results window to make room for the new type of data, and this makes it very
> cumbersome. We could of course do all the manual counting using the line
> tool, but that would take much more time and we have a lot of samples.
>
> I can think of two solutions to this: 1. Make ImageJ accept several types of
> measurements in the Results window at the same time. 2. Let the point
> selection tool (the cross) have a selectable size so that it can function as
> a size reference when doing the point selection. I have looked for plugins
> or macros that could help us with this but haven't found any. Is anybody
> aware of something that could solve this problem? Is there somthing I am
> doing wrong? Could a plugin be made to solve this?
>
> Background:
> We are an ICES working group that is responsible for making egg-production
> estimates for mackerel and horse-mackerel in the North Sea and Norwegian Sea
> from mature pre-spawning ovaries. The egg-production estimates are then
> compared to the amount of pelagic eggs of these species that is actually
> found (using plankton net) during the spawning season. From this the total
> spawner biomass is calculated, and this again makes an important basis for
> deciding the fishing quotas for these species.
>
> thank you
>
> anders
>
> ..................................
> Anders Thorsen
> Dr. Scient.
> Senior Scientist
> Recruitment Biology and Behaviour
> Institute of Marine Research
> Nordnesgaten 50
> P. O. Box 1870 Nordnes
> N-5817 Bergen, Norway
> Phone: [phone]
> Fax: [phone]
> E-mail: [hidden email]
>
>  

--
passerelle antivirus du campus CNRS de Montpellier
--

[Robert Baer]

You may find Kurt De Vos's cell-counter plugin
(http://rsb.info.nih.gov/ij/plugins/cell-counter.html) helpful for what you
are trying to do.  It gives you the possibility of counting and marking
multiple types of objects in an image.

HTH,
Rob


----- Original Message -----
From: "Thorsen, Anders" <[hidden email]>
To: <[hidden email]>
Sent: Monday, October 16, 2006 5:53 AM
Subject: point selection and counting procedure

> Hi,
>
> We are counting oocytes from whole mount pictures of fish ovaries. We are
> counting a lot of samples so our counting procedure should be as efficient
> as possible. The majority of the oocytes (the dark vitellogenic ones) are
> counted easily using the analyze particle function. However, after the
> particle analysis has been run there is still some oocytes that have not
> been counted. Mainly this is the previtellogenic oocytes that are
> transparent and difficult to threshold on. These oocytes have to be
> counted
> manually on the picture. I find using the point selection tool is in many
> cases efficient, however we should only count oocytes that are larger than
> 150 µm. That means that oocytes around this threshold value has to be
> measured using the line tool to decide wether it should be counted or not.
> Since this is a new type of measurement ImageJ then want to empty the
> results window to make room for the new type of data, and this makes it
> very
> cumbersome. We could of course do all the manual counting using the line
> tool, but that would take much more time and we have a lot of samples.
>
> I can think of two solutions to this: 1. Make ImageJ accept several types
> of
> measurements in the Results window at the same time. 2. Let the point
> selection tool (the cross) have a selectable size so that it can function
> as
> a size reference when doing the point selection. I have looked for plugins
> or macros that could help us with this but haven't found any. Is anybody
> aware of something that could solve this problem? Is there somthing I am
> doing wrong? Could a plugin be made to solve this?
>
> Background:
> We are an ICES working group that is responsible for making egg-production
> estimates for mackerel and horse-mackerel in the North Sea and Norwegian
> Sea
> from mature pre-spawning ovaries. The egg-production estimates are then
> compared to the amount of pelagic eggs of these species that is actually
> found (using plankton net) during the spawning season. From this the total
> spawner biomass is calculated, and this again makes an important basis for
> deciding the fishing quotas for these species.
>
> thank you
>
> anders
>
> ..................................
> Anders Thorsen
> Dr. Scient.
> Senior Scientist
> Recruitment Biology and Behaviour
> Institute of Marine Research
> Nordnesgaten 50
> P. O. Box 1870 Nordnes
> N-5817 Bergen, Norway
> Phone: [phone]
> Fax: [phone]
> E-mail: [hidden email]
>

[Thorsen, Anders]

Hi,

Thank you for the idea about the ROI manager. I have now found a very good
solution I think: I make a circular ROI that have the diameter of my lower
size threshold. When the particle analysis is fisnished I use the ROI manger
to establish the circlar ROI. I have configured my mouse so that one of the
knobs now works as a "click lock". When I click on the ROI with the click
lock it follows my mouse afterwards. When the ROI is over a cell that is
larger than the ROI I run a "measure and label" macro. I can then continue
to the next cell without any mouse clicking. Another good think about this
is that the data registered from the measure and label command using a
circular ROI is the same data type as from preceding the particle analysis.
This means that all the data can be registred in the same Results window.

thank you!

anders

> From: Volker Bäcker <[hidden email]>
> Reply-To: ImageJ Interest Group <[hidden email]>
> Date: Mon, 16 Oct 2006 13:37:49 +0200
> To: <[hidden email]>
> Subject: Re: point selection and counting procedure
>
> Hi,
> just an idea: Can't you use the Roi Manager to save your point selection?
> You select points. Now you want to measure a distance, so you add your
> point selection to the roi manager. You make the line selection and
> write the
> distance to the results table. Then you restore your point roi from the
> roi manager and you continue to add points (using shift).
> When you finished, you measure the point roi and get your count as the
> number of rows in the results table.
> You could as well put line rois of a specific size into the roi manager
> and use them as reference.
>
> This is more a workaround then a real solution.
> Maybe it would be possible to adapt the new version of the cell counter
> plugin
> http://rsb.info.nih.gov/ij/plugins/cell-counter.html
> to allow for length measurements during the process of counting.
>
> Hope this helps,
> Volker
> Thorsen, Anders a écrit :
>> Hi,
>>
>> We are counting oocytes from whole mount pictures of fish ovaries. We are
>> counting a lot of samples so our counting procedure should be as efficient
>> as possible. The majority of the oocytes (the dark vitellogenic ones) are
>> counted easily using the analyze particle function. However, after the
>> particle analysis has been run there is still some oocytes that have not
>> been counted. Mainly this is the previtellogenic oocytes that are
>> transparent and difficult to threshold on. These oocytes have to be counted
>> manually on the picture. I find using the point selection tool is in many
>> cases efficient, however we should only count oocytes that are larger than
>> 150 µm. That means that oocytes around this threshold value has to be
>> measured using the line tool to decide wether it should be counted or not.
>> Since this is a new type of measurement ImageJ then want to empty the
>> results window to make room for the new type of data, and this makes it very
>> cumbersome. We could of course do all the manual counting using the line
>> tool, but that would take much more time and we have a lot of samples.
>>
>> I can think of two solutions to this: 1. Make ImageJ accept several types of
>> measurements in the Results window at the same time. 2. Let the point
>> selection tool (the cross) have a selectable size so that it can function as
>> a size reference when doing the point selection. I have looked for plugins
>> or macros that could help us with this but haven't found any. Is anybody
>> aware of something that could solve this problem? Is there somthing I am
>> doing wrong? Could a plugin be made to solve this?
>>
>> Background:
>> We are an ICES working group that is responsible for making egg-production
>> estimates for mackerel and horse-mackerel in the North Sea and Norwegian Sea
>> from mature pre-spawning ovaries. The egg-production estimates are then
>> compared to the amount of pelagic eggs of these species that is actually
>> found (using plankton net) during the spawning season. From this the total
>> spawner biomass is calculated, and this again makes an important basis for
>> deciding the fishing quotas for these species.
>>
>> thank you
>>
>> anders
>>
>> ..................................
>> Anders Thorsen
>> Dr. Scient.
>> Senior Scientist
>> Recruitment Biology and Behaviour
>> Institute of Marine Research
>> Nordnesgaten 50
>> P. O. Box 1870 Nordnes
>> N-5817 Bergen, Norway
>> Phone: [phone]
>> Fax: [phone]
>> E-mail: [hidden email]
>>
>>  
>
> --
> passerelle antivirus du campus CNRS de Montpellier
> --

[Thorsen, Anders]

Hi,

thank you for drawing my attention to cell-counter macro. I think I have
found another nice solution to this particular counting problem (see reply
to Bäcker). However, the cell counter macro will be very useful for us in
some of the other work we do.

cheers

anders

> From: Robert Baer <[hidden email]>
> Reply-To: ImageJ Interest Group <[hidden email]>
> Date: Mon, 16 Oct 2006 08:42:27 -0500
> To: <[hidden email]>
> Subject: Re: point selection and counting procedure
>
> You may find Kurt De Vos's cell-counter plugin
> (http://rsb.info.nih.gov/ij/plugins/cell-counter.html) helpful for what you
> are trying to do.  It gives you the possibility of counting and marking
> multiple types of objects in an image.
>
> HTH,
> Rob
>
>
> ----- Original Message -----
> From: "Thorsen, Anders" <[hidden email]>
> To: <[hidden email]>
> Sent: Monday, October 16, 2006 5:53 AM
> Subject: point selection and counting procedure
>
>
>> Hi,
>>
>> We are counting oocytes from whole mount pictures of fish ovaries. We are
>> counting a lot of samples so our counting procedure should be as efficient
>> as possible. The majority of the oocytes (the dark vitellogenic ones) are
>> counted easily using the analyze particle function. However, after the
>> particle analysis has been run there is still some oocytes that have not
>> been counted. Mainly this is the previtellogenic oocytes that are
>> transparent and difficult to threshold on. These oocytes have to be
>> counted
>> manually on the picture. I find using the point selection tool is in many
>> cases efficient, however we should only count oocytes that are larger than
>> 150 µm. That means that oocytes around this threshold value has to be
>> measured using the line tool to decide wether it should be counted or not.
>> Since this is a new type of measurement ImageJ then want to empty the
>> results window to make room for the new type of data, and this makes it
>> very
>> cumbersome. We could of course do all the manual counting using the line
>> tool, but that would take much more time and we have a lot of samples.
>>
>> I can think of two solutions to this: 1. Make ImageJ accept several types
>> of
>> measurements in the Results window at the same time. 2. Let the point
>> selection tool (the cross) have a selectable size so that it can function
>> as
>> a size reference when doing the point selection. I have looked for plugins
>> or macros that could help us with this but haven't found any. Is anybody
>> aware of something that could solve this problem? Is there somthing I am
>> doing wrong? Could a plugin be made to solve this?
>>
>> Background:
>> We are an ICES working group that is responsible for making egg-production
>> estimates for mackerel and horse-mackerel in the North Sea and Norwegian
>> Sea
>> from mature pre-spawning ovaries. The egg-production estimates are then
>> compared to the amount of pelagic eggs of these species that is actually
>> found (using plankton net) during the spawning season. From this the total
>> spawner biomass is calculated, and this again makes an important basis for
>> deciding the fishing quotas for these species.
>>
>> thank you
>>
>> anders
>>
>> ..................................
>> Anders Thorsen
>> Dr. Scient.
>> Senior Scientist
>> Recruitment Biology and Behaviour
>> Institute of Marine Research
>> Nordnesgaten 50
>> P. O. Box 1870 Nordnes
>> N-5817 Bergen, Norway
>> Phone: [phone]
>> Fax: [phone]
>> E-mail: [hidden email]
>>