Abstract
Methods to measure the performance of computer programs to support medical decision making have been described for situations with a single correct diagnosis (Habbema et al,1976,1978,1981). However, patients frequently have more than one diagnosis. Furthermore, many diagnoses are not ’definite’ and carry some degree of uncertainty. These aspects are even more likely to exist for the type of diagnostic problems for which such systems may be consulted. Therefore, parameters are needed expressing the correctness of systems that offer multiple and weighted diagnoses. These parameters should not only provide insight in the performance of a knowledge base to developers, but also must be easy to understand for the intended users of the system.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Adlassnig K-P, Scheithauer W. Performance evaluation of medical expert systems using ROC curves. Comput Biomed Res 1989; 22: 287–93
Bernelot Moens HJ. Development and testing of a computer program for the diagnosis of joint pain. (In Dutch) Proceedings MIC 88 1988; 201–6
Bernelot Moens HJ, Van der Korst, JK, Kingsland LC. Evaluatie van AI/RHEUM: de gegevens die de arts-gebruiker nodig heeft. (in Dutch) Proceedings MIC 90 1990; 77–85
DeDombal FI’, Horrocks JC. Use of Receiver Operating Characteristic (ROC) curves to evaluate computer confidence threshold and clinical performance in the diagnosis of appendicitis. Meth Inf Med 1978; 17: 157–61
Habbema JDF. Models for diagnosis and detection of combinations of diseases. in: DeDombal FT & Gremy F Eds. Decision Making and Medical Care. 1976; 399–411
Habbema JDF, Hilden J, Bjerregaard B. The measurement of performance in probabilistic diagnosis (parts 1–3). The problem, descriptive tools, and measures based on classification matrices. Meth Inf Med 1978; 17: 217–46
Habbema JDF, Hilden J, Bjerregaard B. The measurement of performance in probabilistic diagnosis (parts 4 and 5). Meth Inf Med 1981; 20: 80–100
Indurkhya N, Weiss S. Models for measuring performance of medical expert systems. Artificial Intelligence in Medicine 1989; 1: 61–70
Kingsland LC. The evaluation of medical expert systems: experience with the AI/Rheum knowledge based consultant system in rheumatology. Proc SCAMC IX, Washington, DC IEEE Comp. Soc. Press 1985; 292–5
Lundsgaarde HP. Evaluating Medical Expert Systems. Soc Sci Med 1987; 24: 805–19
Porter JF, Kingsland LC, Lindberg DAB et al. The AI/RHEUM knowledge-based computer consultant system in rheumatology. Performance in the diagnosis of 59 connective tissue disease patients from Japan. Arthritis Rheum 1988; 31: 219–26
Robertson WO. Quantifying the meanings of words. JAMA 1983; 249: 2621–2
Shapiro AR. The evaluation of clinical predictions: a method and initial application. N Engl J Med 1977; 296: 1509–14
Spiegelhalter DJ. Statistical methodology for evaluating gastrointestinal symptoms. Clin Gastroenterology 1985; 14: 489–515
Toogood JH. What do we mean by “usually”? Lancet 1980; is 1094
Wyatt J, Spiegelhalter DJ. Evaluating medical decision-aids: what to test, and how ? Med Inform 1990; 15: 205–17
Zagoria RJ, Reggia JA. Transferability of medical decision support systems based on Bayesian classification. Med Decis Making 1983; 3: 501–9
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bernelot Moens, H.J., van der Korst, J.K. (1991). Measuring performance of a Bayesian decision support system for the diagnosis of rheumatic disorders. In: Stefanelli, M., Hasman, A., Fieschi, M., Talmon, J. (eds) AIME 91. Lecture Notes in Medical Informatics, vol 44. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-48650-0_12
Download citation
DOI: https://doi.org/10.1007/978-3-642-48650-0_12
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54144-8
Online ISBN: 978-3-642-48650-0
eBook Packages: Springer Book Archive