Computational and noncomputational clinical information processing by computer

Abstract

Computers have been used in their computational as well as noncomputational modes for the scientific study of the clinical decision making. As computational tools, computers serve as storage devices for hospital file data so that these data can be analyzed statistically for a large variety of epidemiological and diagnostic purposes. As noncomputational devices, computers are functioning as knowledgable medical teachers that interface with students who can query the system. In this mode, computers are sometimes used as consultants to clinicians who need information on technical topics. Noncomputational computers are also used to simulate the inferences of clinicians. Byproducts of these simulations are computer programs that become formal statements or theories of the decision processes under investigation.

Artificial intelligence expert systems for clinical diagnosis: Are they worth the effort?

“…Then, on the basis of this information they can learn, from their mistakes, to form better clinical inferences. Of particular interest here is the suggestion by Fryback and Erdman (1979) and Kleinmuntz (1982), among others, that computers can be useful in this regard. They enable assessment and follow-up information on a large number of patients to be stored, and can therefore provide clinicians with accurate statistics, such as the relative frequencies of various diseases.…”

“…Alternatively, the simple linear model may be abandoned as dehumanizing in favor of an approach which focuses more directly on experts' decision-making processes per se. Whether or not this is an example of cognitive conceit, it is precisely the approach taken by Kleinmuntz (1968Kleinmuntz ( ,1975Kleinmuntz ( ,1982 when he had both clinical psychologists and neurologists explain each step of their respective decisionmaking processes. Moreover, it is this emphasis on how decisions are reached, together with the length of time involved in eliciting and editing 'talk aloud' protocols, that prompted Leal and Pearl (1977) to devise an interactive computer program capable of capturing the ways in which decision makers structure their knowledge in order to make their professional decisions.…”

“…They make the diagnostic decisions and computers simply provide realistic, unbiased feedback on those decisions. Kleinmuntz (1982) also sees clinicians in the primary role with computers, as described so far, serving strictly as diagnostic aids that calculate Beta-weights and probabilities and are essentially occupied with number crunching.…”

“…There is one problem, however. Psychological researchers and, in fact, people in general are simply not prepared to have clinicians' diagnostic role replaced by a machine (Kleinmuntz, 1982). Clinical intuition and experience seem to count for something.…”

“…Therefore, the validity of the regression coefficients can be checked from time to time and adjusted to reflect changes when necessary (Kleinmuntz, 1982). SecBehnviornl Science.…”