This Data file contains experiments used to validate the theory explained in the submitted paper entitled :

Spurious Phase Correction in Rapid Metabolic Imaging

written by :

Nour EL SABBAGH(1,2,3), Carine CHASSAIN(1,2,3), Hélène RATINEY(4), Guilhem PAGES(1,2), Jean-Marie BONNY(1,2).

(1) INRAE, UR QuaPA, F-63122 Saint-Gènes-Champanelle, France

(2) AgroResonance, INRAE, 2018. Nuclear Magnetic Resonance Facility for Agronomy, Food and Health, doi: 10.15454/1.5572398324758228E12

(3) Université Clermont Auvergne, CHU, CNRS, Clermont Auvergne INP, Institut Pascal, F-63000 Clermont-Ferrand, France.

(4) Université Lyon 1, CNRS, Inserm, INSA Lyon, UCBL, CREATIS, UMR5220, U1294, F-69621 Villeurbanne, France

Abstract:

Rapid multi-shot MRSI sequences, like IDEAL spiral, require the acquisition of several datasets using optimized sampling in the time domain to reconstruct metabolite maps. Each unitary scan consists of a selective slice (2D) or slab (3D) excitation followed by an evolution time and then the acquisition of the spatially encoded signal. It is critical that the phase variation during the evolution time for each scan is only dependent on chemical shifts. In this paper, we highlight that the phase may depend on either the excitation (fe) or the reception (fr) frequency depending on where the switch from the first to the second frequency is performed in the pulse sequence. To override this, we propose two types of solutions. The first one is based on post-processing signal correction while the second one consists of performing the frequency switch at an appropriate position in the pulse sequence. In order to illustrate this issue, we used the IDEAL spiral method applied to a 13C phantom.

The first application, which was published in the paper, was a general case for multiple metabolites (multiple spins) :

    * 13C Metabolic Imaging using IDEAL spiral.

The second one, which was not published, deals with only 1 metabolite (1 spin) :

    * 1H High Resolution B0 map.

Each application has a folder containing:

    * a READ ME file for a detailed description,

    * a Data folder containing the experiments dataset,

    * a Matlab folder containing the script used.