[Tubetk-developers] GradientBasedAngleOfIncidenceImageFilter

[Andinet Enquobahrie]

Yep.

itkGradientBasedAngleOfIncidenceImageFilter is implementation of the
first part of Equation 3 ( See attached ) i.e diffusive reflection
angle

http://tubetk.org/gitweb?p=TubeTK.git;a=blob;f=Base/USTK/itkGradientBasedAngleOfIncidenceImageFilter.txx;h=55c4c68198058e5eb0a66d9ab84ba2d5ee4a71e9;hb=0f3610b7dbd08a4bbd9c4c6e895040051fab3d17

Our goal is to ultimately implement Equation 6 (Attached also) which
includes both reflected and transmitted ultrasound wave

Andinet


On Wed, Aug 29, 2012 at 9:13 PM, Matt McCormick
<matt.mccormick at kitware.com> wrote:
> On Wed, Aug 29, 2012 at 8:46 PM, Stephen Aylward
> <stephen.aylward at kitware.com> wrote:
>> Shouldn't the vessels be bright on all edges aligned with the gradient
>> normal - regardless of the sign.   That is, shouldn't the near and far
>> edges of the vessels be bright - not just the near ones, as shown?
>> And this applies to every object's edges (liver, etc.).
>>
>> For example, see
>> http://www.medison.ru/uzi/eho398.htm
>> http://www.medison.ru/uzi/eho240.htm
>> and
>> google "ultrasound of liver" and look at the images.
>
> Yes, correct.
>
> This filter is one step to getting there.  Other steps to take:
>
> The ultrasound signal intensity is the result of the product of the
> angle of incidence with a term that depends on the magnitude of the
> acoustic impedance gradient (e.g. Wein 2007 Eqn. 3).  So, we still
> have to compute the result.
>
> The near and far walls will be bright then after taking the absolute
> value.  It seems more appropriate to take the absolute value at a
> later stage so that the angle of incidence filter can possibly be used
> for other purposes that take advantage of the sign.
>
> We can also isolate the sparse "informative points", organ boundaries
> and vessels walls, where these type of specular reflections will occur
> from large scatterers (Szabo section 9.2, 11-Chap09.pdf), possibly
> using Andinet's sheetness filter.
>
> Thanks,
> Matt
>
>>
>> s
>>
>> On Wed, Aug 29, 2012 at 6:32 PM, Matt McCormick
>> <matt.mccormick at kitware.com> wrote:
>>> Hi Andinet,
>>>
>>> I merged the GradientBasedAngleOfIncidenceImageFilter.  Renderings of
>>> the test outputs are attached.  I'll create another filter to
>>> calculate the reflected intensity and fix any dashboard issues that
>>> crop up tomorrow.  We are now up to 600 tests -- woohoo! :-)
>>>
>>> Thanks,
>>> Matt
>>>
>>> _______________________________________________
>>> Tubetk-developers mailing list
>>> Tubetk-developers at tubetk.org
>>> http://public.kitware.com/cgi-bin/mailman/listinfo/tubetk-developers
>>>
>>
>>
>>
>> --
>>
>> ==============================
>> Stephen R. Aylward, Ph.D.
>> Senior Director of Operations - North Carolina
>> Kitware, Inc. - North Carolina Office
>> http://www.kitware.com
>> stephen.aylward (Skype)
>> (919) 969-6990 x300



-- 
Andinet Enquobahrie, Ph.D.
Technical Lead
Kitware, Inc.

http://www.kitware.com
(919) 969-6990 x311
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Equation3.png
Type: image/png
Size: 9689 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/tubetk-developers/attachments/20120830/72fcfb80/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Equation6.png
Type: image/png
Size: 14873 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/tubetk-developers/attachments/20120830/72fcfb80/attachment-0005.png>