Abstract #1976
Dont use relative Cramer Rao lower bounds for elimination of low quality data!
Roland Kreis 1 and Sreenath Pruthviraj Kyathanahally 1
1
Depts. Radiology and Clinical Research,
University Bern, Bern, Switzerland
It is demonstrated that the use of relative Cramer-Rao
Lower Bounds (CRLB) for quality filtering can easily
lead to wrong conclusions either to believe in metabolic
alterations where there are none, or to miss significant
effects. This is illustrated with synthetic examples. In
addition, consequent application of rejection of data
with high relative error prevents the clinical use of
MRS in any disease leading to very low metabolite
levels. CRLB are valuable to judge the trust one can
have in a MRS-based measurement, but it has to be judged
either as an absolute value or relative to the
normal/control levels.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
