Extreme value statistics for novelty detection in biomedical data processing

Extreme value theory (EVT) is a branch of statistics which concerns the distributions of data of unusually low or high value i.e. in the tails of some distribution. These extremal points are important in many applications as they represent the outlying regions of normal events against which we may wish to define novel events. The use of such novelty detection approaches is useful for analysis of data for which few exemplars of some important class exist, for example in medical screening. It is shown that a principled approach to the issue of novelty detection may be taken using extreme value statistics.