Retrieving accurate relaxometric information from low signal-to-noise ratio 23Na MRI performed in vivo

Sodium is the second best nucleus in living systems MRI after proton, but it is the first one concerning its sensitivity to the electrical configuration of the environment, due to its quadrupolar relaxation properties. Retrieving accurate relaxation data from low signal-to-noise ratio (SNR) experiments implies using post-processing procedures in order to fit correctly the decaying relaxation curves. The conventional image reconstruction method is introducing a positive bias due to the rician noise contribution. Phasing the images restores its initial Gaussian characteristic. Linear prediction (LP) forward of the decays improves the SNR of the last points and calculates missing points. An objective fitting method, based on singular value decomposition (SVD) allows us to determine the number of exponentials existing in the decay and its accurate fitting parameters. Sodium ions' spin–spin relaxation holds great potential to become an endogeneous marker for tumor diagnosis and for therapy response, since different regions of a mouse tumoral liver bear distinguishable relaxation parameters as compared to control.