Submitted by bpittman. on 2006-01-03 08:59.
It is common in MRA to display the 3D results by a Maximum Intensity Projection (MIP) for 'white blood' imaging, and a minimum Intensity Projection (mIP) for 'black blood' imaging.
Submitted by bpittman. on 2005-12-29 13:18.
Lee Friedman suggested that EPI images should be smoothed to a given smoothness prior to time series analysis.
Submitted by bpittman. on 2005-12-28 09:46.
Gaussian blurring, as in AFNI's edt_blur.c functions, is easily implemented using FFTs and/or convolution. But if there is a mask involved, and we don't want to include data from outside the mask, what to do?
Submitted by bpittman. on 2005-12-23 08:56.
Estimation of the physiological noise without external monitoring should be possible.
Submitted by bpittman. on 2005-12-22 16:29.
It would be nice to have a family of simple high-order multi-dimensional warping functions that were easily invertible -- ideally, the family would contain its own inverse functions.
Submitted by bpittman. on 2005-12-21 09:43.
Registering DWI images for DTI calculations has the problem that the different gradient weightings produce markedly different images. The result is that the Gx-direction image looks quite a lot different from the Gy-weighted image (say). This poses problems for standard registration methods.
Submitted by bpittman. on 2005-12-21 09:29.
Measuring the gradient fields in MRI is actually tricky, without using some sort of external fiducial object. Here's one way to avoid this problem.
Submitted by bpittman. on 2005-12-20 14:25.
An attempt to unify the various theories of DWI (q-space and tensors). Not complete, and maybe not correct.
Submitted by bpittman. on 2005-12-20 14:01.
Don't we wish.