[A comparative evaluation of computerized diagnostic decisional support in gastroenterology in relation to typical clinical pictures].

On the basis of the results obtained in a previous perspective comparative study aimed to compare the diagnostic accuracy of a computerized diagnostic decisional support in gastroenterology with that of non-specialist physicians, the Authors have carried out a retrospective study, based on the same series and on the same results, but disaggregating the global comparisons according to the feature of the clinical presentations. The latter were therefore classified by independent gastroenterologists as "typical", "atypical", or "borderline", and for each group the difference of diagnostic accuracy between program and physicians was again evaluated. Even if the program proved more accurate in all groups, the greatest difference was observed in the "borderline" group which the Authors claim to include the kind of presentation which more than any other needs a decisional aid. This should suggest a possible useful implementation of the system in daily clinical practice.