This paper deals with the reconstruction of T1-T2 correlation spectra in nuclear magnetic resonance relaxometry. The ill-posed character and the large size of this inverse problem are the main difficulties to tackle. While maximum entropy is retained as an adequate regularization approach, the choice of an efficient optimization algorithm remains a challenging task. Our proposal is to apply a truncated Newton algorithm with two original features. First, a theoretically sound line search strategy suitable for the entropy function is applied to ensure the convergence of the algorithm. Second, an appropriate preconditioning structure based on a singular value decomposition of the forward model matrix is used to speed up the algorithm convergence. Furthermore, we exploit the specific structures of the observation model and the Hessian of the criterion to reduce the computation cost of the algorithm. The performances of the proposed strategy are illustrated by means of synthetic and real data processing.
The aim of this work was to study the spin–spin (T2) relaxation components of one hard cheese and three soft cheeses to characterize fat and water states. NMR signals were measured at 6 °C with a 0·47 T NMR device. The transverse relaxation decay was fitted using the Marquardt method. The T2 relaxometric behaviour of the cheeses under consideration was characterized by four relaxation components. To understand the chemical composition of each NMR component, we studied anhydrous milk fat extracted from each cheese analysed. At 6 °C, the fat was 60% crystalline. In cheese, the solid fat was found mainly in the shorter relaxation component with a T2 of 17 μs. The intensity of the NMR relaxation with a T2 > 1 ms was explained by the amount of water, liquid fat and proteins, and the associated relaxation time varied as a function of the process used. The composition of each relaxation component was confirmed by the temperature effect and the influence of the fat content on the NMR cheese signal. NMR relaxometry was able to provide information on water behaviour (i.e. the quantity and level of interactions with proteins) and on the solid[ratio ]liquid ratio of anhydrous milk fat in the cheese.