Subscribe to RSS
DOI: 10.1055/s-0038-1634827
Problem Area Formation as an Element of Computer Aided Diagnosis: A Comparison of Two Strategies Within Quick Medical Reference (QMR)[*]
The authors wish to thank Janine E. Janosky, Ph. D., for her valuable help with the statistical analysis. Ou thanks also to Richard Bankowitz, Ma D. and Nunzia Guise, M. D., for then assistance with study conceptualization and design. Dr. Miller’s work has been supported by a National Library of Medicine Research Career Development Award K04-LM-00084. Dr. Fred E. Masarie, Jr. has co-developed the QMR program in association with Dr. Miller. The QMR Knowledge Base has been developed and maintained at the University of Pittsburgh by Drs. J. D. Myers, R. A. Miller and F. E. Masarie, Jr.Publication History
Publication Date:
07 February 2018 (online)
Abstract
INTERNIST-I’s use of “partitioning” to group related diagnoses into problem areas (for competitive consideration and elimination during case analysis) is felt to be the source of many of its strengths as well as some of its weaknesses. QMR, INTERNIST-I’s successor program, embodies a homology function which can act as an alternative to the partitioner for problem area formation. This study undertakes a comparison of the problem areas generated by the INTERNIST-I partitioning algorithm, the QMR homology function, and expert clinicians; it finds the correlation to be poor. The authors then discuss another method of problem area formation which might better mimic a human clinician and provide an alternative approach in diagnostic computer-aided decision making.
Key-Words
Computer-Assisted Medical Diagnosis - Problem Area Formulation - Quick Medical Reference (QMR) - Evaluation of Decision-Making Strategy* QMR and Quick Medical Reference arc Registered Trademarks of the University of Pittsburgh.
-
References
- 1 Miller RA, McNeil MA, Challinor SM. et al. The INTERNIST-1/QUICK MEDICAL REFERENCE project-status report. West J Med 1986; 145: 816-22.
- 2 Miller RA, Pople HE, Myers JD. INTER-NIST-1, an experimental computer-based diagnostic consultant for general internal medicine. N Engl J Med 1982; 307: 468-76.
- 3 Masarie FE, Miller RA, Myers JD. INTER-NIST-1 properties: Representing common sense and good medical practice in a computerized medical knowledge base. Comput Biomed Res 1985; 18: 458-79.
- 4 Pople HE. Heuristic methods for imposing structure on ill-structured problems: The structuring of medical diagnostics. In: Szolovits P. ed. Artificial Intelligence in Medicine. AAAS Symposium Series. Boulder CO: Westview Press; 1982: 119-85.
- 5 Banks G. Artificial intelligence in medical diagnosis. The INTERNIST/CADUCEUS approach. CRC Crit Rev Med Inf 1986; 01: 23-54.
- 6 Weed LL. Medical records that guide and teach. N Engl J Med 1982; 278 593-600, 652-7.
- 7 Hurst JW, Walker HK. eds. The problem oriented system. New York, NY: Medcom Learning Systems; 1972
- 8 Case Records of the Massachusetts General Hospital (Case 16-1987). N Engl J Med 1987; 316: 1008-17.
- 9 Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977; 33: 363-74.
- 10 Fleiss JL, Nee JC, Landis JR. Large sample variance of kappa in the case of different sets of raters. Psychological Bulletin 1979; 86: 974-7.
- 11 Miller RA, Masarie FE. The demise of the “Greek Oracle” model for medical diagnostic systems. Meth Inform Med 1990; 29: 1-2.
- 12 Parker RC, Miller RA. Creation of realistic appearing simulated patient cases using the INTERNIST-I/QMR knowledge base and interrelationship properties of manifestations. Meth Inform Med 1989; 28: 346-51.
- 13 Miller RA, Masarie FE. Use of the Quick Medical Reference (QMR) program as a tool for medical education. Meth Inform Med 1989; 28: 340-5.
- 14 Eddy DM, Clanton CH. The art of diagnosis - Solving the Clinicopathological exercise. N Engl J Med 1982; 306: 1263-8.