ImageJ

Does this make sense? Measuring zebrafish axons.

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

5 messages Options

Holloway, Michael

Does this make sense? Measuring zebrafish axons.

Anyone have any experience/thoughts on this? I need an automated, or
semi-automated method of measuring zebrafish motor neuron axon length (see,
for instance: http://www.ncbi.nlm.nih.gov/pubmed/22462669). The methods
out there all seem to use area (pixel counts) as a proxy for length. The
major problem with this is that the intensity of fluorescence can't be kept
constant across samples. To some degree, one sample's axons will appear
fatter than the next. One software solution (
http://www.sciencedirect.com/science/article/pii/S0165027006000112) uses a
plugin, FeatureJ, to adjust thresholds in the region of interest and define
edges by calculating the smallest eigen value before counting pixels.
Should I expect this to correct, by some degree, for inconsistent label
fluorescence intensity? If not, why not just subtract background and
measure density?

The other problem is that a zebrafish tail has depth, and motor axons grow
along both sides. A projection would be expected to hide axons behind its
neighbor. I'm thinking that I should rotate the 3d image till there's
maximum axon area.

Thanks for any comments.

Mike Holloway

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

Krs5

Re: Does this make sense? Measuring zebrafish axons.

Dear Mike

Not sure I totally understand the question as it seems you want to measure the neuron length but then talk about "axons will appear fatter" what might suggest you are after volume information. If you want length maybe have a look at Simple Neurite Tracer (http://fiji.sc/Simple_Neurite_Tracer) which works also on 3D data sets.

Best wishes

Kees

Dr Ir K.R. Straatman
Centre for Core Biotechnology Services
University of Leicester
http://www.le.ac.uk/biochem/microscopy/home.html

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Holloway, Michael
Sent: 17 September 2013 19:13
To: [hidden email]
Subject: Does this make sense? Measuring zebrafish axons.

Anyone have any experience/thoughts on this? I need an automated, or semi-automated method of measuring zebrafish motor neuron axon length (see, for instance: http://www.ncbi.nlm.nih.gov/pubmed/22462669). The methods out there all seem to use area (pixel counts) as a proxy for length. The major problem with this is that the intensity of fluorescence can't be kept constant across samples. To some degree, one sample's axons will appear fatter than the next. One software solution (
http://www.sciencedirect.com/science/article/pii/S0165027006000112) uses a plugin, FeatureJ, to adjust thresholds in the region of interest and define edges by calculating the smallest eigen value before counting pixels.
Should I expect this to correct, by some degree, for inconsistent label fluorescence intensity? If not, why not just subtract background and measure density?

The other problem is that a zebrafish tail has depth, and motor axons grow along both sides. A projection would be expected to hide axons behind its neighbor. I'm thinking that I should rotate the 3d image till there's maximum axon area.

Thanks for any comments.

Mike Holloway

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

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

Holloway, Michael

Re: Does this make sense? Measuring zebrafish axons.

Neurite Tracer and similar Imagej plugins don't actually measure length.
They count pixels and give an area measure. If two identical axons have
different fluorescent intensities the pixels counted will be different. The
question then is if this known source of error can be worked with, or is
there another automated way of doing it.

Thanks,
Mike Holloway
On Sep 18, 2013 4:43 AM, "Straatman, Kees R. (Dr.)" <[hidden email]>
wrote:

> Dear Mike
>
> Not sure I totally understand the question as it seems you want to measure
> the neuron length but then talk about "axons will appear fatter" what might
> suggest you are after volume information. If you want length maybe have a
> look at Simple Neurite Tracer (http://fiji.sc/Simple_Neurite_Tracer)
> which works also on 3D data sets.
>
> Best wishes
>
> Kees
>
>
> Dr Ir K.R. Straatman
> Centre for Core Biotechnology Services
> University of Leicester
> http://www.le.ac.uk/biochem/microscopy/home.html
>
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Holloway, Michael
> Sent: 17 September 2013 19:13
> To: [hidden email]
> Subject: Does this make sense? Measuring zebrafish axons.
>
> Anyone have any experience/thoughts on this? I need an automated, or
> semi-automated method of measuring zebrafish motor neuron axon length (see,
> for instance: http://www.ncbi.nlm.nih.gov/pubmed/22462669). The methods
> out there all seem to use area (pixel counts) as a proxy for length. The
> major problem with this is that the intensity of fluorescence can't be kept
> constant across samples. To some degree, one sample's axons will appear
> fatter than the next. One software solution (
> http://www.sciencedirect.com/science/article/pii/S0165027006000112) uses
> a plugin, FeatureJ, to adjust thresholds in the region of interest and
> define edges by calculating the smallest eigen value before counting pixels.
> Should I expect this to correct, by some degree, for inconsistent label
> fluorescence intensity? If not, why not just subtract background and
> measure density?
>
> The other problem is that a zebrafish tail has depth, and motor axons grow
> along both sides. A projection would be expected to hide axons behind its
> neighbor. I'm thinking that I should rotate the 3d image till there's
> maximum axon area.
>
> Thanks for any comments.
>
> Mike Holloway
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

Volker Baecker

// Spam /// Re: Does this make sense? Measuring zebrafish axons.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Hello Michael,
are you sure that Simple Neurite Tracer does not measure length? The
length of the path it shows seems to be the sum of the Euclidean
distances between the points of the tracing. The "analyze skeleton"
that you can run from the SNT seems to measure the length basically in
the same way.
If you do not trust it I guess you could rather easily calculate the
length from the exported points in the tracing (the .swc file) yourself.
But maybe I'm missing something?
Best regards
Volker

Holloway, Michael:

> Neurite Tracer and similar Imagej plugins don't actually measure
> length. They count pixels and give an area measure. If two
> identical axons have different fluorescent intensities the pixels
> counted will be different. The question then is if this known
> source of error can be worked with, or is there another automated
> way of doing it.
>
> Thanks, Mike Holloway On Sep 18, 2013 4:43 AM, "Straatman, Kees R.
> (Dr.)" <[hidden email]> wrote:
>
>> Dear Mike
>>
>> Not sure I totally understand the question as it seems you want
>> to measure the neuron length but then talk about "axons will
>> appear fatter" what might suggest you are after volume
>> information. If you want length maybe have a look at Simple
>> Neurite Tracer (http://fiji.sc/Simple_Neurite_Tracer) which works
>> also on 3D data sets.
>>
>> Best wishes
>>
>> Kees
>>
>>
>> Dr Ir K.R. Straatman Centre for Core Biotechnology Services
>> University of Leicester
>> http://www.le.ac.uk/biochem/microscopy/home.html
>>
>>
>> -----Original Message----- From: ImageJ Interest Group
>> [mailto:[hidden email]] On Behalf Of Holloway, Michael Sent:
>> 17 September 2013 19:13 To: [hidden email] Subject: Does
>> this make sense? Measuring zebrafish axons.
>>
>> Anyone have any experience/thoughts on this? I need an
>> automated, or semi-automated method of measuring zebrafish motor
>> neuron axon length (see, for instance:
>> http://www.ncbi.nlm.nih.gov/pubmed/22462669). The methods out
>> there all seem to use area (pixel counts) as a proxy for length.
>> The major problem with this is that the intensity of fluorescence
>> can't be kept constant across samples. To some degree, one
>> sample's axons will appear fatter than the next. One software
>> solution (
>> http://www.sciencedirect.com/science/article/pii/S0165027006000112)
>> uses a plugin, FeatureJ, to adjust thresholds in the region of
>> interest and define edges by calculating the smallest eigen value
>> before counting pixels. Should I expect this to correct, by some
>> degree, for inconsistent label fluorescence intensity? If not,
>> why not just subtract background and measure density?
>>
>> The other problem is that a zebrafish tail has depth, and motor
>> axons grow along both sides. A projection would be expected to
>> hide axons behind its neighbor. I'm thinking that I should
>> rotate the 3d image till there's maximum axon area.
>>
>> Thanks for any comments.
>>
>> Mike Holloway
>>
>> -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>> -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>
> -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

Holloway, Michael

Re: // Spam /// Re: Does this make sense? Measuring zebrafish axons.

I'm afraid I missed that entirely. I didn't look far enough into the
Neurite Tracer documentation and tutorial. At least I got it working.
Thanks greatly.

Thanks,
Mike Holloway

On Wed, Sep 18, 2013 at 8:36 AM, Volker Baecker
<[hidden email]>wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
>
> Hello Michael,
> are you sure that Simple Neurite Tracer does not measure length? The
> length of the path it shows seems to be the sum of the Euclidean
> distances between the points of the tracing. The "analyze skeleton"
> that you can run from the SNT seems to measure the length basically in
> the same way.
> If you do not trust it I guess you could rather easily calculate the
> length from the exported points in the tracing (the .swc file) yourself.
> But maybe I'm missing something?
> Best regards
> Volker
>
> Holloway, Michael:
> > Neurite Tracer and similar Imagej plugins don't actually measure
> > length. They count pixels and give an area measure. If two
> > identical axons have different fluorescent intensities the pixels
> > counted will be different. The question then is if this known
> > source of error can be worked with, or is there another automated
> > way of doing it.
> >
> > Thanks, Mike Holloway On Sep 18, 2013 4:43 AM, "Straatman, Kees R.
> > (Dr.)" <[hidden email]> wrote:
> >
> >> Dear Mike
> >>
> >> Not sure I totally understand the question as it seems you want
> >> to measure the neuron length but then talk about "axons will
> >> appear fatter" what might suggest you are after volume
> >> information. If you want length maybe have a look at Simple
> >> Neurite Tracer (http://fiji.sc/Simple_Neurite_Tracer) which works
> >> also on 3D data sets.
> >>
> >> Best wishes
> >>
> >> Kees
> >>
> >>
> >> Dr Ir K.R. Straatman Centre for Core Biotechnology Services
> >> University of Leicester
> >> http://www.le.ac.uk/biochem/microscopy/home.html
> >>
> >>
> >> -----Original Message----- From: ImageJ Interest Group
> >> [mailto:[hidden email]] On Behalf Of Holloway, Michael Sent:
> >> 17 September 2013 19:13 To: [hidden email] Subject: Does
> >> this make sense? Measuring zebrafish axons.
> >>
> >> Anyone have any experience/thoughts on this? I need an
> >> automated, or semi-automated method of measuring zebrafish motor
> >> neuron axon length (see, for instance:
> >> http://www.ncbi.nlm.nih.gov/pubmed/22462669). The methods out
> >> there all seem to use area (pixel counts) as a proxy for length.
> >> The major problem with this is that the intensity of fluorescence
> >> can't be kept constant across samples. To some degree, one
> >> sample's axons will appear fatter than the next. One software
> >> solution (
> >> http://www.sciencedirect.com/science/article/pii/S0165027006000112)
> >> uses a plugin, FeatureJ, to adjust thresholds in the region of
> >> interest and define edges by calculating the smallest eigen value
> >> before counting pixels. Should I expect this to correct, by some
> >> degree, for inconsistent label fluorescence intensity? If not,
> >> why not just subtract background and measure density?
> >>
> >> The other problem is that a zebrafish tail has depth, and motor
> >> axons grow along both sides. A projection would be expected to
> >> hide axons behind its neighbor. I'm thinking that I should
> >> rotate the 3d image till there's maximum axon area.
> >>
> >> Thanks for any comments.
> >>
> >> Mike Holloway
> >>
> >> -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >>
> >> -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >>
> >
> > -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >
>
> -----BEGIN PGP SIGNATURE-----
>
> iQEcBAEBCgAGBQJSOZ47AAoJENcVU956o58NB1EH/2a/eft1wERTCJBK1QrnSni9
> qz+sElkOkeYzfFIKoAqleBFpegIm1wirEqcBjoORuaew9tpgs3FNQSndBrpSUKHM
> KIQnyUpuyV8KzpopqVhtXt9ehT5vifYCo1Chq9PNVgkKyAlhcZez7l5fEYhucoX3
> BBjwk3fxK7NLI0In6Gln5FEPu+E+/mTxXWdVjgJXkbn9+e48gVZ1nZDprPIJ3mje
> 0LKHlpdJi54CWNZFRU6OiiXQeFkOSUJvCKu85A8nqbMtC9yJQ9LvE8ZqszntBCLe
> aV9OzZHvC22yt/p2K2bpR8SL9lOe2YfXmLM88V9prQNf7Y5APA1iCDg/Zh5IgEs=
> =eOYO
> -----END PGP SIGNATURE-----
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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