| Re: Three questions [message #11458 is a reply to message #11406] |
Mon, 13 April 1998 00:00   |
David Foster
Messages: 341 Registered: January 1996
|
Senior Member |
|
|
biomedical wrote:
>
> Thanks a lot for your help. It's really a great help
>
> My purpose is to do the auto segmentation of the medical images,
> you must have a lot of experience. The original images are not binary,
> they are all kinds of medical image formats, some are color photos (for
> wound healing). I want to output the auto-segmentation results in
> either ROI(lines connected by points) or binary tiff images (
> encoded by run-length).
If you're working with medical images you might want to check out
my library of functions/programs available at:
ftp://bial8.ucsd.edu pub/software/idl/share
The README file lists the files with short descriptions. There should
be routines you could use. Also, we do semi-automated segmentation
of brain MRI scans, using a regression analysis which is seeded
with samples chosen by the user. If you want, email me and I can
send you more info on our technique.
I would recommend saving the segmentation results as either:
1) ROI: list of indices into array; this will allow
faster access than a list of vertices or lines would.
2) GIF images
Note that you can use POLYFILLV() with the RUN_LENGTH keyword to
return the indices of a polygon in run-length form.
Dave
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~
David S. Foster Univ. of California, San Diego
Programmer/Analyst Brain Image Analysis Laboratory
foster@bial1.ucsd.edu Department of Psychiatry
(619) 622-5892 8950 Via La Jolla Drive, Suite 2240
La Jolla, CA 92037
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~
|
|
|
|