Resampling algorithms for diagnostic radiology.

The use of digital image resampling can be highly suitable for diagnostic radiology. It can be applied not only to obtain zoom effects but also to solve various problems related to the comparison of images of different origin (NMR, CT, NM). The mathematical approach to this kind of algorithm, based on rational assumptions, can be very wide and diversified. In this work the principles of the stochastic and of the analytical methods have been analyzed. In particular, a complete mathematical treatment of the integral analytical methods, which are both the most theoretically interesting and analytically complex, has been performed. The practice of these techniques has to be determined each time from the analysis of the specific problem to which they are to be applied in the routine, and after a retrospective analysis of the results obtained.