hello spm-ers,
please excuse that my message is slightly off topic.
could anyone recommend some good info (book, paper, website, etc) on
error propagation?
specifically what I am interested is the same question in two forms:
(1) suppose i observe a certain signal intensity in a voxel. I would
like to figure out how sensetive that value is to hardware instability.
how much will my signal vary if for example the flip angle is in error
by 2%.
(2) or the other way around. suppose I notice that the signal in some
voxel (say a phantom) varies by 5% from measurement to measurement. I
would like to be able to say that my flip angle is probably not in error
by more than 2% because that would lead to a variation in the voxel
signal with a higher % of variation.
thanks in advance,
zoltan
|