Validation of the medical expert system RENOIR

Christine Hernandez, Joan J. Sancho, Miguel A. Belmonte, Carles Sierra, Ferran Sanz

Research output: Contribution to journalArticleResearchpeer-review

17 Citations (Scopus)

Abstract

RENOIR is an expert system developed to assist the diagnosis of 37 diseases of connective tissue and inflammatory arthropathies. Precise diagnosis of rheumatic diseases implies great uncertainty and there is no gold standard with which to comparethe expert system output. To overcome this problem a set of clinical cases was submitted to RENOIR andits diagnoses were compared with those of clinicians. Medical records or 81 patients with rheumatic diseases were interpreted by RENOIR and by 12 clinicians at three different expertise levels in rheumatology. Distances between the likelihoods of the 37 considered diseases provided by clinicians and RENOIR were computed as a disagreement measure. Mahalanobis distance was used to correct the collinearity between the possibilities of each pair of diseases. Using the resulting matrices of distances between experts, cluster analyses were carried out to classify RENOIR among human experts. Greater differences between RENOIR and clinicians than among clinicians themselves were not found. © 1994 by Academic Press, Inc.
Original languageEnglish
Pages (from-to)456-471
JournalComputers and Biomedical Research
Volume27
Issue number6
DOIs
Publication statusPublished - 1 Jan 1994

Fingerprint

Dive into the research topics of 'Validation of the medical expert system RENOIR'. Together they form a unique fingerprint.

Cite this