[vtkusers] negative values in CT data

[Richard A. Ketcham]

CT imagery is characteristically scaled so that image values match 
Hounsfield Units, which is a 12-bit scale with air corresponding to -1000 
and water equal to 0, and bone in the region of 800.  It was all done in 
the 1960's, when imaging technologies were young and zero did not seem so 
important, and the standards have stuck.  A simple conversion is to just 
add 1000 or so (or perhaps a bit more, to account for noise in the air).

Rich Ketcham


At 10:39 AM 6/30/00 -0400, David E. Jones wrote:
>There are a lot of vtk users who do volume visualization.  I do
>this myself for data collected with microtomography.  There is a
>characteristic of the reconstructed data which I don't understand.
>It's this:  there are a lot of negative values in the data.  I've
>attached a histogram which shows this.  The histogram is symmetric
>which suggests that the negative values are meaningful.  I've also
>attached a picture of one of the slices with level=.2 and
>window=1.4 using the units on the historgram scale.
>
>Is anyone familiar with the occurence of negative values and what
>to do with them ?  I know that the negative values need to be
>eliminated to use volume rendering.  In the past I've set them
>to be 0 but I'm not sure this is correct.
>
>Thanks in advance,
>Dave
>--
>David E Jones
>Du Pont Central Research
>Experimental Station, Bldg 320
>Wilmington, DE 19880-0320

______________________________
Dr. Richard Ketcham
Research Associate
High-Resolution X-Ray Computed Tomography Facility
Department of Geological Sciences
University of Texas
Austin, TX   78712-1101
Office: (512) 471-0260       Fax: (512) 471-9425
Email: richk at maestro.geo.utexas.edu
http://www.ctlab.geo.utexas.edu