The localisation of the activation response from an fMRI experiment depends on obtaining reference scans which highlight the anatomy of the subjects brain. For this purpose an inversion recovery image can be particularly beneficial, since the contrast in the grey and white matter regions of the brain can be altered by the use of a magnetisation inversion r.f. pulse and an appropriate delay before imaging (see section 2.4.1).
The pulse sequence diagram for a standard inversion recovery experiment is shown in Figure 4.14. At 3.0 Tesla the values of the inversion time, TI, required are 400 ms and 1200 ms for white matter null, and grey matter null respectively.
Acquiring the reference set, with a suitable number of repeats to gain sufficient signal to noise, can take several minutes. Much of this time is taken up by the inversion time delay. The fast inversion recovery sequence uses this time to invert other slices, before going back to image them, as shown in Figure 4.15.
A constant repetition rate is chosen which is an integer multiple of the required inversion time. For example, to obtain grey matter null images with an inversion time of 1200 ms, four slices are inverted at a repetition rate of 300 ms. A comparison of images obtained using the conventional IR sequence with those obtained using the fast IR sequence shows no difference between the two, however for a typical set of white and grey matter images the fast sequence takes just over 5 minutes compared to 12 minutes for the standard IR method. The main limit to the speed of the fast IR acquisition is the rate of inversion pulses that it is safe to use, because of SAR considerations.
This reduction in time that the subject stays in the scanner helps to keep the subject motivated, and minimises discomfort, which in turn will reduce motion artefact.