MRQuantif 'Results': curve and concentrations

The options

This is a very important part of the software. The different options available influence the value of the results:

• LIC unit: You can choose to use the unit of expression of hepatic iron concentration per gram of liver (dry weight) by selecting either μmol/g if you work with hepatologists or mg/g if you work with hematologists. There is a factor 18 between these two values: 1 mg/g = 18 μmol/g. This explains why the upper limit of the normal initially fixed at 2 mg/g is also 36 μmol/g.

• T2 *: the software proposes different methods of calculation of T2 * (only for the liver):
  • T2 * optimal: the software chosen T2 * that seems optimal taking into account the values ​​obtained and the correlation coefficient at the points used.
  • T2 * basic: this is the calculation of the T2 * taking all the points of the liver, without taking into account the phase and without taking into account any background noise measured or calculated . It is understood that the measurement can be disturbed by the presence of fat or distorted in the event of significant collapse of the signal.
  • T2 * offset: it's the same thing but considering that there is an offset corresponding to the background noise. This unknown is calculated by the formula. In practice, the offset value obtained is sometimes false and far from the measured value of the background noise.
  • T2 * in: this is the calculation of the T2 * basic but taking only the points of the liver close to the phase.
  • T2 * out: this is the calculation of the T2 * basic but taking only the points of the liver close to the phase opposition.
  • T2 * subtraction: it's the same as the T2 * basic, but subtracting the background signal from the liver signal (calculated by offset or measured). This method often leads to a reduction in T2 *.
  • T2 * truncation: it's the same as the T2 * basic but stopping to take into account the points of the hepatic signal from the first point which is equal to or less than background noise. It is certainly the most logical method but it can also be faulted if the signal is collapsed.
  • T2 * extrapolation: it's the same as the T2 * truncation but adding fictitious points adapted to the shape of the curve (personal idea not yet evaluated).


• LIC T2 *: the software allows you to choose the conversion formula that gives the hepatic iron concentration (CHF or LIC in English) from T2 * or R2 * (T2 * = 1/R2 * and vice versa). At 1.5T, there are several published references giving slightly different results (Garbowski 2014, Henninger and Wood). If you work with a formula you can select it, otherwise by default and in a recommended way for now choose instead that of Garbowski. At 3T, there is only one reference.


• LIC SIR: the software also allows you to choose the algorithm for calculating the hepatic iron concentration (CHF or LIC in English) from the liver-to-muscle ratio. At 1.5T, there are two published references giving different results (Alustiza and Gandon). If you work with a formula you can select it, otherwise by default and in a recommended way for now choose instead of Alustiza. Indeed the method we propose probably increases the average iron overload. At 3T, there is only, up to know, one reference.
  


• Select: finally it is possible to define if one wishes to prioritize one of the two methods, T2 * or SIR, or if the system has to take into account the T2 * in priority and to use the SIR method if the measured T2 * is too short compared to the shortest TE used (T2 * <2 TE min) or if the correlation between the T2 * curve and the measured signal is not high enough (r2 <0.8).