conversion from HU to pixel intensity.

hello,can ny1 help me out wit d following:- i want pixel-intensity equivalent of the following:- -1000HU,-400HU. i'm pretty new to matlab,i want equivalent values which fall under the range 0-255.thanks in advance..

Answer 1

[Walter Roberson]

See http://www.ctlab.geo.utexas.edu/overview/index.php#anchor3-3 "Although CT values should map linearly to the effective attenuation coefficient of the material in each voxel, the absolute correspondence is arbitrary. "

In other words, there is no fixed conversion between pixel values and CT numbers (HU).

Also note the word "should" -- looking around at some papers, it appears to me that there are conditions under which the CT numbers do not map linearly.

[Walter Roberson]

I _think_ you edited your question to say specifically that you wanted values in the range 0 to 255. It doesn't matter, though, as my response still stands.

The correspondence between HU and pixel intensities has to be calibrated for the equipment in use. It also varies according to what the experimenter was interested in looking at.

At least one of the articles I looked at said explicitly that a "windowing" technique was used: that is, the experimenter knows calibration points for the equipment, and then from the measureable range, selects a HU range of interest to what is being looked for, and that HU range is mapped to the 0-255 or 0-4095 pixel intensity range.

Your question is similar to presenting us with a thermometer containing an unknown fluid and asking us what height of liquid correspond to what temperature: a question that cannot be answered without calibrating or else detailed examination of the properties of the liquid.

[basheer jan]

thanks it was quite helpful.