Expert diagnosis in tall stature: edits 1.1 diagnostic software efficacy

Abstract

EDITS (Expert Diagnosis In Tall Stature) expert system version 1.1 was evaluated by making it diagnose sets of clinical and laboratory criteria published as case reports in medical scientific literature, and analyzing the results obtained. The case reports were selected by means of the PubMed database. Processing the input data yielded a list of the top five most-probable disorders, ranked in the order of likelihood along with their probability. 21 cases of 12 diseases were processed. In 14 cases out of 21 (66.67%) the referral diagnosis was included in the top five most probable diagnoses listed by the expert system. In 64.29% of all the cases when the correct diagnosis was put on the top five list, the correct diagnosis ranked first. The efficacy of EDITS 1.1 is in line with that demonstrated by SYNDROC, POSSUM, OMIM and the Expert System for Differential Diagnosis of Hereditary Neuromuscular Diseases in Children. EDITS 1.1 diagnostic software proved to be a useful practical tool for differential diagnosis of disorders that may be manifested by the tall stature syndrome.