[tomjen3]

I do - and someday you may do so as well.

One of the things they are trying to GPU accelerate is improved MRI scannings - imagine taking images for every 1mm and then stitch it together so that you get a 3d model of the area which the doctors can look at from every angle.

Unfortunately that takes way, way too much computer power to be feasible today.

[nodata]

Why do the images need to be stitched together?

[z303]

Guessing here but so they can be turned into a polygon model for easier display, yes you can directly volume render the data but say a 1024 x 1024 x 1024 cube even at a byte per voxel would be a large amount of memory, most of which will be empty

I'm guessing OpenCL is being used to accelerate Marching cubes or some other similar algorithm

http://en.wikipedia.org/wiki/Marching_cubes

This post by Smash of Fairlight http://directtovideo.wordpress.com/2011/05/03/numb-res/ talks about some of the reason why you want to do this but in a very different context

[tomjen3]

Because otherwise you have a bunch of images and not a 3D object.

But this is far from the only thing people use it for. Basically it is used, or might be used, anywhere else you would use a cluster.

Here [http://www.youtube.com/watch?v=14ZSXCqOML0] is a real-time rendering of the Julia set (in 4D, so you are really watching the projection).