The invention described provides amethod of quantitatively evaluating one or more of thelikelihood. severity and progression of a disease frommedical images comprising processing medical images ofa test subject to derive one or more feature space valuescharacteristic of a disease-dependent image attributes,comparing the feature space values to those of a previouslyestablished database from medical images of knownhealth} and known diseased subjects, wherein the comparingis based on feature space values that best discriminatebetween health and diseased subjects, summing aweighted distance of discriminant feature space values ofthe test subject to those of at least one of the mean featurespace value of the healthy subjects and the mean featurespace value of the diseased subjects, and providing fromthe summing a single number which is indicative of atleast one of disease likelihood, severity and progression.
