vessel diameter

[Teodora-Adriana Barbacaru]

Hi,

I need to estimate the mean vessel diameter of some tortuous tumor
vessels from vessel sections that do not look elliptical at all.

I did this in several ways:
the smaller side of the bounding rectangle
the diameter of the inscribed circle centered at the centroid
the breadth perpendicular to the feret diameter
minor axis of the best fitting ellipse

By the way, how does ImageJ fit the ellipse? Inscribed ellipse,
enclosing ellipse, ellipse with the same surface, having the feret of
the structure as major axis? What is the algorithm for this fit?

All methods result in different values. Does anyone know which is the
most appropriate method to determine the vessel diameter? Do you know
any references discussing advantages and disadvantages of every method?


Adriana Perles

[Harry Parker]

Hi,

What is the reason behind finding the "diameter" of an irregularly shaped region? The method you need depends on your answer.
Thinking
as a physicist and engineer: Is it to derive the cross section area, or
to determine the perimeter? Both impact the amount of blood flow to the
tumor in opposite directions.
Measuring the area and perimeter of an
irregular enclosed region are better defined. Then you can derive what
the diameter would be from either of these measures. I.e : What is the
diameter of a circle with the same area?  What is the diameter of a
circle with the same perimeter? Of course these give 2 different
answers. The difference is a measure of the circularity is the region.
 
--  
Harry Parker  
Senior Systems Engineer  
Digital Imaging Systems, Inc.

----- Original Message ----
From: Teodora-Adriana Barbacaru <[hidden email]>
To: [hidden email]
Sent: Wednesday, April 4, 2007 6:59:28 AM
Subject: vessel diameter

Hi,

I need to estimate the mean vessel diameter of some tortuous tumor
vessels from vessel sections that do not look elliptical at all.

I did this in several ways:
the smaller side of the bounding rectangle
the diameter of the inscribed circle centered at the centroid
the breadth perpendicular to the feret diameter
minor axis of the best fitting ellipse

By the way, how does ImageJ fit the ellipse? Inscribed ellipse,
enclosing ellipse, ellipse with the same surface, having the feret of
the structure as major axis? What is the algorithm for this fit?

All methods result in different values. Does anyone know which is the
most appropriate method to determine the vessel diameter? Do you know
any references discussing advantages and disadvantages of every method?


Adriana Perles





 
____________________________________________________________________________________
Food fight? Enjoy some healthy debate
in the Yahoo! Answers Food & Drink Q&A.
http://answers.yahoo.com/dir/?link=list&sid=396545367

[Dimiter Prodanov]

Dear Adriana,

this is quite common problem. You can also try another stereological
quantity:
"objects per unit length" (in your cases vessel wall). you project
serial parallel test lines with a random orientation (of the first one)
and then you count the
number of hits with the vessel walls. the you divide this to the total
length of the test lines in the image
In this way you avoid the whole discussion of sizes and
the relevance of the elliptical fits.
best

D Prodanov

[Albert Cardona]

Adriana,

If what you have is a stack, then use Kai's Volume Viewer plugin to
rotate it in the 3D space until you see one section of the tube as
transversal as possible, then push 'save image' and measure such
diameter in the opened snapshot image.

http://rsb.info.nih.gov/ij/plugins/volume-viewer.html

Albert