Electrical impedance tomography is a recently developed imaging technique, with which images of the internal impedance of the subject can be rapidly collected with rings of external ECG-type electrodes. Its advantages are that it is fast, inexpensive, portable, and is very sensitive to physiological changes which affect the electrical impedance properties. Set against this is a relatively poor spatial resolution and the ill-posed nature of the procedure for reconstructing images, such that small errors due to instrumentation translate into large errors in images. The work of our group has been to adapt existing EIT designs for the demanding application of imaging changes in the brain due to conditions like stroke, epilepsy or normal physiological brain activity; the difficulty is that the skull is resisted and diverts current so that the signal-to-noise ratio is low. Signal processing to try to optimise the signal-to-noise ratio is an essential part of the method in areas such as data acquisition, data pre-processing and image reconstruction.