[Danielle Graveron]

In jMRUI, the noise standard deviation can be estimated from the residue, the last points of the raw signal or from a known value in case of simulations. Users define the value used in the estimation of the standard deviation on metabolite amplitudes (AMARES, QUEST, etc.) in the Results window.

Danielle Graveron
Universite Claude Bernard - LYON 1

[Danielle Graveron]

For background handling, QUEST exploits the fact that macromolecule and lipid signals decay rapidly in the time-domain, so the major part of the background signal is in the very first points of the signal. Strategic truncation of these initial samples can separate the metabolite part from any fast decaying parts and thus decorrelate metabolite and nuisance parameters.

How to choose the number of the pseudo-truncated points?

The lipid (if not too huge) and macromolecule signals must have decayed into the noise by sample point Ntrunc. This number depends on the magnetic field strength and the sampling interval. For rat brain MRS data at 7T, I use data 20 points but it is probably not the best value for you case.

Danielle Graveron-Demilly

Danielle Graveron
Universite Claude Bernard - LYON 1

[Author]

Posts