A new method for diagnosing multiple diseases in Large medical decision sup
port systems based on causal probabilistic networks is proposed, The method
is based on characteristics of the diagnostic process that we believe to b
e present in many diagnostic tasks, both inside and outside medicine. The d
iagnosis must often be made under uncertainty, choosing between diagnoses t
hat each have small prior probabilities, but not so small that the possibil
ity of two or more simultaneous diseases can be ignored, Often a symptom ca
n be caused by several diseases and the presence of several diseases tend t
o aggravate the symptoms. For diagnostic problems that share these characte
ristic, we have proposed a method that operates in a number of phases: in t
he first phase only single diseases are considered and this helps to focus
the attention on a smaller number of plausible diseases, In the second phas
e, pairs of diseases are considered, which make it possible to narrow down
the held of plausible diagnoses further. In the following phases, larger su
bsets of diseases are considered,
The method was applied to the diagnosis of neuromuscular disorders, using p
revious experience with the so-called MUNIN system as a starting point. The
results showed that the method gave large reductions in computation time w
ithout compromising the computational accuracy in any substantial way, It i
s concluded that the method enables practical inference in large medical ex
pert systems based on causal probabilistic networks.