AMIDE / News: Recent posts

The most notable change in this release, is that the MS-Windows version finally contains all the same features as the Linux version. Also, the windows binary no longer requires an external GTK runtime package.

See the ChangeLog for a full listing of changes.

The underlying version of GTK was upgraded from 2.2 to 2.10 in this release, and a fair amount of underlying code has been changed. Although there should be no real visible changes, due to the large number of changes you may want to upgrade with caution.

AMIDE version 0.8.15 has just been released. The biggest feature of this release is much improved DICOM support via the use of the DCMTK library. See the Changelog for the complete list of changes.

One caveat in this release to those who use Concorde files. The handling of the branching fraction has been changed in the (X)MedCon library (it was ignored previously), so the generated numbers for these files will be different.

New version is out, see the change log for details if you're interested.

MS Windows users will be especially interested in this release, as it fixes the "halt" bug with ROI drawing that occurs on Windows system after installing the RPC update (the fix for the lovesan worm).

After quite a wait, a version of AMIDE is finally available for microsoft windows. Be warned, that the windows version of AMIDE should be considered beta at best, and your mileage may vary.

New version is out, with the usual mix of new features and bug fixes. Check the ChangeLog for the whole story.

I've started maintaing a Mac OS X port also (using fink). Instructions on installing AMIDE under OS X are now in the FAQ on the web page.

The dependency on mpeg_encode has been replaced by a dependency on libfame (can be found at http://fame.sf.net\). Mpeg_encode is still on the web page, as the 0.6 branch still needs it for movie generation.

You'll notice that there's now 2 versions of AMIDE. The old 0.6.x branch (currently at 0.6.4) is GTK 1.2 based, and is now considered stable. I will no longer be adding to this branch, I'll just be correcting any remaining bugs that happen to pop up in the future.

The 0.7.x branch is a complete rewrite of essentially all the underlying code, with huge improvements in terms of maintainability of the code base. A lot of underlying cruft has been stripped out, and replaced by a fairly elegant object oriented hierachy in the same spirit as the GTK libraries. For a preliminary glimpse of the new architecture, check out the API section of the web page. 0.7.x is also the port to the new GTK 2.0/GNOME 2.0 libraries. I'm sure there's a couple bugs left to be found, but it's been pretty stable for me so far.... read more

Many changes in this one, the most noteable addition being the ability to draw fiducial reference points and perform rigid body alignments. Have fun.

gcc 2.95.2 started generating bad code when optimizations were used with this release. I've tried 2.95.3, 3.0.1, and 3.0.2, and the only version that will generate correct code is 3.0.2, so for now on the recommended gcc version is 3.0.2. If you want to use a previous version, see the FAQ on the web page for suggestions.

The vast majority of the changes in this release are invisible to the user. The internal data representation has been drastically altered to allow for non-floating point data sets to be stored in memory. This alteration allows a 50% decrease in memory use and disk space for most data sets.

One configuration/compilation issue, libamide is no longer built by default (it's now a configuration option).... read more

AMIDE 0.4.2 is now out. You will also notice that new volpack, libecat, and mpeg_encode packages are present. This is because all of these packages have been restructured so that they'll compile cleanly on Mac OS X (and other unixes besides Linux).

The only package that AMIDE requires that doesn't compile cleanly is xmedcon, and that should be fixed in the next version (needs new config files).... read more

AMIDE 0.3.2 is out. The only two changes from 0.3.1 are that (supposedly) it'll now compile on big endian machines, and that a couple more view parameters are saved in the study files. Now that I'm finally done with my medical boards, work should start progressing again on the program.

In other news, I'll be presenting a poster on AMIDE at the Society of Nuclear Medicine meeting in Toronto. The poster is #827 - "AMIDE: a completely free system for medical imaging data analysis". My scheduled time for presenting is 8:00 AM till 9:30 AM (gag) on Monday morning (June 25th).

New version is out (0.2.3) which includes volume rendering support using the volpack volume rendering library. The volume rendering is currently a little buggy, but works well enough to get the screenshots shown on the homepage. Numerous small bug fixes to the base system have also been included in this release.

Amide -0.2 is out. Handling of several volumes at once is now correctly implemented, so registration between different images/modalities is possible. The ability to import raw data files has also been added, along with numerous smaller changes and bug fixes which I won't bother mentioning. Volume rendering is working, but I haven't yet incorporated it into a release. Expect it soon.