[New approaches to computer-assisted diagnosis of rheumatologic diseases].

UNLABELLED

Since the 1960s, several knowledge-based systems for computer-assisted diagnosis in radiology have been developed. The great majority of these tools has been implemented as off-line systems. This requires interaction with the system solely for the purpose of consultation and therefore interrupts the radiologist's work flow. This and inadequate man-machine interfaces may have inhibited the routine clinical use of such systems. The goal of this paper is to describe the current research toward the development of the on-line expert system Cadiag-4/Rheuma-Radio. The underlying fundamentals of the system design, including client/server architecture, communication interfaces, and fuzzy set theory and fuzzy logic as methods for knowledge representation and interference, are presented.

METHODS

In radiology today, computers are routinely used to acquire radiological images in hospital and radiology information systems (HIS/RIS) and picture archiving and communication systems (PACS). In our approach, we make use of pre-existent sources of information to build an expert system that minimizes the interaction between radiologists and the computer. To handle uncertainty and vagueness of medical knowledge, fuzzy set theory and fuzzy logic are used. Given data of a specific case, a deductive inference procedure combines the observed radiological signs, establishes confirmed and excluded diagnoses as well as diagnostic hypotheses, and provides explanations for these conclusions. Furthermore, proposals for confirmation or exclusion of diagnostic hypotheses are offered.

RESULTS

For evaluation purposes, an early prototype of Cadiag-4/Rheuma-Radio was tested on radiological disorders of the hip joint related to rheumatological diseases. Twenty radiological cases were used as test cases, reaching a diagnostic accuracy of about 80%.

CONCLUSION

The first results are acceptable and encourage further work to cover the whole area of rheumatologically relevant radiological signs and diagnoses. Furthermore, research into the development of user-oriented data acquisition tools will be carried out.