Renal T2-mapping with a T2-preparation sequence: comparison of signal models

Abstract

A recent consensus paper recommends a mono-exponential signal model to determine T2-values from a T2-preparation sequence. However, this assumes complete signal recovery after each readout, and therefore necessitates long acquisition times. In this study, we compare the mono-exponential model against a forward modelling approach which is also accurate with incomplete recovery. Simulations, phantom data and repeatability data in healthy volunteers show the forward model is significantly more accurate and allows for a 7-fold reduction in acquisition time with a negligible cost in T2 precision.