Knee MRI Models

[Dr. Mike]

I was recently asked the following question by a member, and I thought I would share the answer so others in the community could benefit.

---------------------

Dear Dr. Mike,

 

My thesis focuses on knee biomechanics and I was hoping I could pick your brain in this regard. I have to obtain MRI scans of a knee which will be later segmented and the model will be eventually used for finite element analysis. Since I have very limited knowledge in this, I was wondering what parameters (such as sequence, weighting, resolution, slice thickness) should be maintained during an MR procedure so as to obtain the best contrast between various soft tissues which will make the segmentation easier?

 

Thanks for your help!

-----------------------

 

I recently completed an MRI knee segmentation (image below). The best way to do this is to use a high resolution (sub 1mm slice thickness) T1 or proton density series of the knee. If you want to exclude the fat, you can do a fat-saturation sequence, but my understanding is that it is hard to get thin slices with fat-sat sequences. I used Slicer's segmentation module to segment the individual structures. If you want to put the entire knee back together, you can add the structures together after they are all segmented.

 

Hope this helps!

Dr. Mike