On 16-Jan-09 0:28, gregjohn wrote:
> Twice I've had MRI's in the past year.  Both times I tried to strike up a
> conversation about getting my own copy of the data in a 3D format.  Both times
> I struck out. Neither of them knew what DXF was.  One of them said that the end
> product report featured a "rotatable" view of the examined structure, "but it's
> not 3D."
> 
> Of course, I just wanted my own copy to play in 3D, but I'm wondering if I'm
> just using language they don't understand, or if I've totally misunderstood
> what (and WHEN in the analysis process) there is "3D" data.
> 
> Q: Has anyone ever gotten say a DXF file of their knee to play with?  How did
> you get it, and when did you ask?
> 
If by DXF you mean a vector format then no. MRI is in Dicom format, that 
contains bitmapped slices with position and orientation information. If 
you have a regular space stack (general from a CT machine, sometimes 
enough slices in a set from an MRI) then the most logical format to 
convert to is a density file. For triangles you have to apply some other 
software. Easy to do, but not the way the data is normally stored. I do 
have my own torso and heart, not my extremities. I get that sort of data 
by asking the technician to put it on a CD/DVD or on a special server 
that I have access to ;)
The 3D rotatable thing is rendered as an isosurface in real time.

Next time ask for the Dicom files and I may be able to assist in a 
conversion.

Post a reply to this message