New plugin: Skeletonize3D and AnalyzeSkeleton

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

New plugin: Skeletonize3D and AnalyzeSkeleton

Ignacio Arganda Carreras
Dear all,

I just uploaded two plugins for 2D/3D image skeletonization and skeleton
analysis:

http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start

http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:skeletonize3d:start

Please, have a look at them and make me know if you think it would be
interesting to add more measures at the skeleton analysis.

I hope you find them useful.

Cheers!

ignacio

--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

R Heberto Ghezzo, Dr
Salud Ignacio,
Un punto que podrias agregar, creo que lo puedo hacer yo en mi copia, es de contar
los puntos triples, o sea las ramificaciones " branching points".
El numero de puntos triples divided por el area de la figura es un estimador morfometrico
de la longitud del esqueleto.
Yo trabajo en morfometria pulmonar, uso ImageJ para calcular el Lm or 'mean linear
intercept' como una medida de emfisema que es en realidad un estimador de la razon area/volumen.
como nosotros contamos el numero de celulas en las paredes que son positivas para algun marcador
necesitamos un estimador de la longitud de esas paredes, que sea mas facil que trazar las
paredes.
muchas gracias, salud y pesetas ( euros)
[hidden email]
 


-----Original Message-----
From: ImageJ Interest Group on behalf of Ignacio Arganda Carreras
Sent: Mon 11/17/2008 5:44 AM
To: [hidden email]
Subject: New plugin: Skeletonize3D and AnalyzeSkeleton
 
Dear all,

I just uploaded two plugins for 2D/3D image skeletonization and skeleton
analysis:

http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start

http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:skeletonize3d:start

Please, have a look at them and make me know if you think it would be
interesting to add more measures at the skeleton analysis.

I hope you find them useful.

Cheers!

ignacio

--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

Ignacio Arganda Carreras
Hello Heberto (i will write in English so all people in the list can
follow our discussion),

I don't think I completely understood you, how are the "branching
points" different from my junction voxels? Aren't they the same?

AnalyzeSkeleton provides the number of branches, junctions, average
branch length and the number of voxels from every type (end-point,
junction and slab).

Thanks a lot for your feedback!

ignacio




R Heberto Ghezzo, Dr wrote:

> Salud Ignacio,
> Un punto que podrias agregar, creo que lo puedo hacer yo en mi copia, es de contar
> los puntos triples, o sea las ramificaciones " branching points".
> El numero de puntos triples divided por el area de la figura es un estimador morfometrico
> de la longitud del esqueleto.
> Yo trabajo en morfometria pulmonar, uso ImageJ para calcular el Lm or 'mean linear
> intercept' como una medida de emfisema que es en realidad un estimador de la razon area/volumen.
> como nosotros contamos el numero de celulas en las paredes que son positivas para algun marcador
> necesitamos un estimador de la longitud de esas paredes, que sea mas facil que trazar las
> paredes.
> muchas gracias, salud y pesetas ( euros)
> [hidden email]
>  
>
>
> -----Original Message-----
> From: ImageJ Interest Group on behalf of Ignacio Arganda Carreras
> Sent: Mon 11/17/2008 5:44 AM
> To: [hidden email]
> Subject: New plugin: Skeletonize3D and AnalyzeSkeleton
>  
> Dear all,
>
> I just uploaded two plugins for 2D/3D image skeletonization and skeleton
> analysis:
>
> http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start
>
> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:skeletonize3d:start
>
> Please, have a look at them and make me know if you think it would be
> interesting to add more measures at the skeleton analysis.
>
> I hope you find them useful.
>
> Cheers!
>
> ignacio
>
>  


--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

R Heberto Ghezzo, Dr
Ignacio,
triple points are the juncture of 3 and only 3 branches ( with a tolerance of 1-2 pixels.)
cross over of 2 branches, ie "cuadruple" points are not considered.
That was my request to differentiate 'triple points' from 'quadruple' or more points.
Thanks
Heberto



-----Original Message-----
From: ImageJ Interest Group on behalf of Ignacio Arganda Carreras
Sent: Mon 11/17/2008 9:35 AM
To: [hidden email]
Subject: Re: New plugin: Skeletonize3D and AnalyzeSkeleton
 
Hello Heberto (i will write in English so all people in the list can
follow our discussion),

I don't think I completely understood you, how are the "branching
points" different from my junction voxels? Aren't they the same?

AnalyzeSkeleton provides the number of branches, junctions, average
branch length and the number of voxels from every type (end-point,
junction and slab).

Thanks a lot for your feedback!

ignacio




R Heberto Ghezzo, Dr wrote:

> Salud Ignacio,
> Un punto que podrias agregar, creo que lo puedo hacer yo en mi copia, es de contar
> los puntos triples, o sea las ramificaciones " branching points".
> El numero de puntos triples divided por el area de la figura es un estimador morfometrico
> de la longitud del esqueleto.
> Yo trabajo en morfometria pulmonar, uso ImageJ para calcular el Lm or 'mean linear
> intercept' como una medida de emfisema que es en realidad un estimador de la razon area/volumen.
> como nosotros contamos el numero de celulas en las paredes que son positivas para algun marcador
> necesitamos un estimador de la longitud de esas paredes, que sea mas facil que trazar las
> paredes.
> muchas gracias, salud y pesetas ( euros)
> [hidden email]
>  
>
>
> -----Original Message-----
> From: ImageJ Interest Group on behalf of Ignacio Arganda Carreras
> Sent: Mon 11/17/2008 5:44 AM
> To: [hidden email]
> Subject: New plugin: Skeletonize3D and AnalyzeSkeleton
>  
> Dear all,
>
> I just uploaded two plugins for 2D/3D image skeletonization and skeleton
> analysis:
>
> http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start
>
> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:skeletonize3d:start
>
> Please, have a look at them and make me know if you think it would be
> interesting to add more measures at the skeleton analysis.
>
> I hope you find them useful.
>
> Cheers!
>
> ignacio
>
>  


--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

Gabriel Landini
On Monday 17 November 2008 18:58:49 R Heberto Ghezzo, Dr wrote:
> triple points are the juncture of 3 and only 3 branches ( with a tolerance
> of 1-2 pixels.) cross over of 2 branches, ie "cuadruple" points are not
> considered. That was my request to differentiate 'triple points' from
> 'quadruple' or more points. Thanks
 
The Lines8 plugin in the latest version of the Morphology collection can
measure (2D only) skeleton lengths (2 types of lengths) and return the number
of 1..8 point pixels in multiple skeletons on screen. To do this, one needs
to select Redirect to: Connectivity

http://www.dentistry.bham.ac.uk/landinig/software/software.html

I cannot understand, however, how can the number of triple points divided by
the area of the object be a reliable estimate of the skeleton length.

A long skeleton or short without triple points would have a length of 0?

Cheers

Gabriel
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

Ignacio Arganda Carreras
Hello all!

I just uploaded a new version of the plugin with "triple points"
calculation. I hope it's useful!

Here you are:

http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start

Cheers!

ignacio



Gabriel Landini wrote:

> On Monday 17 November 2008 18:58:49 R Heberto Ghezzo, Dr wrote:
>  
>> triple points are the juncture of 3 and only 3 branches ( with a tolerance
>> of 1-2 pixels.) cross over of 2 branches, ie "cuadruple" points are not
>> considered. That was my request to differentiate 'triple points' from
>> 'quadruple' or more points. Thanks
>>    
>  
> The Lines8 plugin in the latest version of the Morphology collection can
> measure (2D only) skeleton lengths (2 types of lengths) and return the number
> of 1..8 point pixels in multiple skeletons on screen. To do this, one needs
> to select Redirect to: Connectivity
>
> http://www.dentistry.bham.ac.uk/landinig/software/software.html
>
> I cannot understand, however, how can the number of triple points divided by
> the area of the object be a reliable estimate of the skeleton length.
>
> A long skeleton or short without triple points would have a length of 0?
>
> Cheers
>
> Gabriel
>
>
>
>  


--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda
Reply | Threaded
Open this post in threaded view
|

Re: New plugin: Skeletonize3D and AnalyzeSkeleton

R Heberto Ghezzo, Dr
Hello All, Ignacio & Gabriel
I typed too fast, what I mean was:
The Length per Volume of a structure in an IUR is estimated by
Lv = 2 Qa i.e. 2 * number of intersections of the structure with the cutting plane.
I work with lung tissue, the triple points in the image are the intersections of
the trihedrals of the alveoli with the cutting plane. If I imagine a very convoluted
line that follows those trihedrals, its length per volume is estimated by the number of triple points.
This measure,[ I think, it is my project] will correlate with the amount of emphysema
[lung destruction and/or alveoli enlargement] better than the usual measure of Lm which is an estimation of "Alveolar Volume"/"Alveolar Wall Area"
Thanks Ignacio, I will try your plugin.
[hidden email]


-----Original Message-----
From: ImageJ Interest Group on behalf of Ignacio Arganda Carreras
Sent: Wed 11/19/2008 5:34 AM
To: [hidden email]
Subject: Re: New plugin: Skeletonize3D and AnalyzeSkeleton
 
Hello all!

I just uploaded a new version of the plugin with "triple points"
calculation. I hope it's useful!

Here you are:

http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start

Cheers!

ignacio



Gabriel Landini wrote:

> On Monday 17 November 2008 18:58:49 R Heberto Ghezzo, Dr wrote:
>  
>> triple points are the juncture of 3 and only 3 branches ( with a tolerance
>> of 1-2 pixels.) cross over of 2 branches, ie "cuadruple" points are not
>> considered. That was my request to differentiate 'triple points' from
>> 'quadruple' or more points. Thanks
>>    
>  
> The Lines8 plugin in the latest version of the Morphology collection can
> measure (2D only) skeleton lengths (2 types of lengths) and return the number
> of 1..8 point pixels in multiple skeletons on screen. To do this, one needs
> to select Redirect to: Connectivity
>
> http://www.dentistry.bham.ac.uk/landinig/software/software.html
>
> I cannot understand, however, how can the number of triple points divided by
> the area of the object be a reliable estimate of the skeleton length.
>
> A long skeleton or short without triple points would have a length of 0?
>
> Cheers
>
> Gabriel
>
>
>
>  


--
Ignacio Arganda-Carreras
Escuela Politécnica Superior
Laboratorio B-408     Phone: (+34) 91 497 2260
Universidad Autónoma de Madrid
Ctra. de Colmenar Viejo, Km. 15
Madrid 28049,  Spain

E-mail: [hidden email]
Website: http://www.ii.uam.es/~iarganda