Information Technology Reference
In-Depth Information
cases by means of a “mechanical apparatus” and to identify relationships between
them. [43]
A brief description of this “first 'computer diagnosis' of disease, in this case
hematology disorders” gives the biographical memoir on Hardy by Arthur B.
Dubois: “The computer consisted of punched cards in a shoe box. Diagnostic crite-
ria had been obtained from a hematology textbook and were wedge-punched at the
edge of each of 26 cards to match the symptoms and laboratory findings of the 26
blood disorders. Knitting needles were run through the holes that corresponded to
the symptoms and laboratory findings of each of 80 patients, matching those to the
diagnostic criteria wedge-punched into the edges of the set of 26 hematology cards.
Shaking the box made the card whose criteria matched those of the patient drop out
of the shoe box to show the diagnosis printed on the hematology card.” [22, p. 13f]
Following intensive collaboration between physicians, mathematicians and elec-
trical engineers, medicine became, to a certain extent, a quantitative science. Vari-
ous approaches to computerized diagnosis emerged in the 1960s and 1970s, using
Bayes rule [90, 98], factor analysis [89], and decision analysis [42]. On the other
hand, artificial intelligence approaches also came into use, e.g., DIALOG (Diag-
nostic Logic) [56] and PIP (Present Illness Program) [52]. These were programs to
simulate the physician's reasoning in gathering information, as well as simulate the
diagnosis using databases in the form of networks of symptoms and diagnoses.
As a next step we should mention the introduction of medical expert systems
shortly after general non-fuzzy expert systems appeared in the 1970s. The first of
these being MYCIN
11
[79], INTERNIST [50] and CASNET (Causal Associational
Networks) [92, 93].
Revisiting the work of Ledley and Lusted [42] we notice how the authors con-
sidered also probabilitic concepts in the “Reasoning Foundations of Medical Diag-
noses”. They argued that in many cases our medical knowledge is not exact but
in the form “If a patient has disease 2, then there is only a certain chance that he
will have symptom 2 - that is, say, approximately 75 out of 100 patients will have
symptom 2. [...] Since 'chance' or 'probabilities' enter into 'medical knowledge',
then chance, or probabilities, enter into the diagnosis itself.” [42, p. 13]
However, six years later, it seemed that Lusted had given up the program to use
methods of exact mathematics in medicine; in his contribution to a volume on
Com-
puters in Biomedical Research
[83] he was agreeing with a very new claim: “Re-
search on medical diagnosis has served to emphasize the need for better methods
of collecting and coding medical information and to demonstrate the inadequacy
of conventional mathematical methods for dealing with biological problems. In a
recent statement Professor L. A. Zadeh (1962) summed up the situation as follows:
...” ThenLusted quoted Zadeh's paragraph that we quoted already in chapter 3.3.1.
The first computer assisted system for medical diagnosis using the theory of
fuzzy sets was the Viennese
C
omputer-
A
ssisted
Diag
nostic System CADIAG-II.
11
MYCIN was written in Lisp as the subject of the doctoral dissertation of Edward Shortliffe
(born 1947). This expert system identified bacteria causing severe infections and it recom-
mended the dosage of antibiotics, depending from the patient's body weight. However, it
was never used in practice.
