Can a patient data base help build a stroke diagnostic expert system?
Daniel B. Hier, Gary D. Atkinson, R. Perline, Howard Hill, M. Evans, B. Desai, W. C. McCormick, L. R. Caplan · Medical Informatics · 1986
The availability of a large database of patient information has been of benefit in the construction of an expert system for stroke care. Such a data base can be used to provide a ready supply of test cases for validating the accuracy of expert systems. The database can provide data that is useful in expert systems that rely upon probabilistic reasoning. A factual database makes it possible to make probabilistic diagnoses based upon comparison with similar prior cases. If the database is large enough, it obviates the need to resort to the assumptions of symptom independence. The database can also be used to derive the empirical likelihoods necessary to apply Bayes' theorem and thus avoid using the guesses of human experts.