Number of Signal Measurements
INCORRECT. Your answer might be correct. 256 is a common image size (and so 256 phase encoding steps are needed; 256 MRI signals). But many different image sizes may be chosen.
Image sizes like 128×128, or 256×256 are chosen to speed the Fourier transform calculation, which is much quicker when an image size is to be N by N, where N is a power of 2. A fast Fourier transform (FFT) may be used instead of a discrete Fourier transform. A regular discrete FT on a symmetric matrix requires N2 calculations, whereas the FFT requires Nlog2N calculations. For a 256×256 matrix, this is a reduction of the number of calculations required by a factor of 32!
Digitisation of a MRI signal. This must be repeated many times (with different phase encoding) to produce just one image. Digitised numbers are represented as greyscale values in this diagram.
CORRECT. A large number of MRI signals must be acquired before it is possible to create an adequate image. This is because the same number of signal measurements are required as there are pixels in the phase encoding direction of the image to be created. For example, for a 256×256 image, 256 phase encoding steps are required.
It is helpful to bear the repetitive nature of MRI signal measurement in mind when learning the basics of MRI. What has just happened to the net magnetisation in a voxel, as part of the measurement of the previous signal, may affect the measurement of the next signal.
INCORRECT. A single MRI signal is frequency encoded because the frequency encoding gradient is switched on as the signal is recorded. We have seen that we cannot frequency encode in two directions at once in an imaging slice. In order to encode position into the second direction in an imaging slice, multiple MRI signals must be generated which have a variety of phase differences between the rows in an image. Only then can we use the different rates of change of phase experienced by each row in an imaging slice, to determine position in that direction.