Iliad and Medical HouseCall: evaluating the impact of common sense knowledge on the diagnostic accuracy of a medical expert system
Journal
Proc Annu Symp Comput Appl Med Care
Publication Date
1995
Pages
742-6
Summary:
- HIT Description: Computerized decision support with artificial intelligence. More info...
- Purpose of Study: Assess the diagnostic accuracy of a computerized medical expert system.
- Years of study: 1994-1995
- Study Design: Case study w/concurrent control
- Outcomes: Other outcome
- Settings: The study used 'textbook' cases for the evaluation.
- Intervention: A refined knowledge base (1995) of a medical expert system with added universal and medical 'common sense' knowledge was compared to an older version (1994) 'pre-common sense'.
- Evaluation Method: Measures of system performance and accuracy.
- Description: The medical expert system used was ÒMedical HouseCallÓ, a consumer health information expert system based on a medical expert system for physicians, ÒIliadÓ.
- Changes in efficiency and productivity: Between the 1994 and the 1995 versions, the irrelevancy score decrease from 21% to 5% (t statistic, p value: 6.027, 0.0001), and the proportion of correct diagnosis increased from 19% to 43% (t statistic, p value: 9.442, 0.0001).