The power of an MRI imager is often characterized by the extent of its static magnetic field. In the hospital sector, this can be a 1.5-Tesla field, sometimes 3T. In research, this level can reach 7T for humans, and soon up to 11.7T. The more elevated this field is, the more the spatial resolution of images can increase. MRI devices also use another electromagnetic field, a radio frequency (RF) field, which becomes inhomogeneous when the static field is amplified. These inhomogeneities cause loss of contrast and shadow zones that appear on images of thighs or abdomens at 3T, as well as on images of brains at 7T.
NeuroSpin researchers at the CEA's Institut Frédéric-Joliot have developed a method called "kT points" to reduce these defects, consisting of introducing small pulses of main field gradients interspersed between short RF pulses. This makes it possible to homogenize the signal and the contrast throughout the entire observed organ.
A study of 50 subjects with various liver pathologies imaged at 3T has demonstrated the interest of this method in a clinical setting. Indeed, with conventional MRI, the relationship between images after and before contrast agent injection does not give the expected signal amplification (Figure 1, top right). On the other hand, the transmission method using kT points leads to the expected cartography of this amplification (bottom right). In a blind evaluation, the radiologists judged the image quality to be either good or very good in 85% of the 20 clinical cases using the kT point method, as compared to only 55% using the manufacturer's conventional parallel transmission method.
Thus, the NeuroSpin team has demonstrated that kT points will make it possible (in routine clinical practice) to smooth out inter-individual differences in examination quality, which will make otherwise unreadable acquisitions interpretable.