Drop ascending T2*/S0 estimates #439
Labels
refactoring
issues proposing/requesting changes to the code which do not impact behavior
T2*/S0 estimation
issues related to the estimation of T2* and S0
Summary
We currently estimate T2* and S0 for all ascending subsets of echoes, starting at two. We then retain only those estimates associated with the value in the adaptive mask (except for voxels in the mask with a value of 1, where we use the T2*/S0 estimates from the first two echoes). This leads to a lot of calculations that are not used, as well as function outputs that no one wants. Also, while there is minimal time cost with the default log linear method, with the curve method calculating all of these unnecessary T2*/S0 estimates can take a while.
Additional Detail
We can limit T2*/S0 calculation to voxels and sets of echoes that will actually be used, as well as drop the ascending estimates from our outputs.
Next Steps
The text was updated successfully, but these errors were encountered: