Medical coding has become an important new industry that has originate
d from the field of medical informatics. Automatic coding of specimens
has emerged as a way of relieving hospitals from the cost of paying p
rofessional coders and for achieving uniform coding for all specimens.
Unfortunately, automatic coding, like manual coding, has numerous pit
falls. Further, the coding algorithms employed by manufacturers of aut
omatic coders are typically proprietary. We have developed a method fo
r automatic coding of pathology reports. Using this public domain auto
coder, we have previously demonstrated that automatic SNOMED coding wa
s superior to manual coding in several measurable categories, includin
g the overall number of codes generated and the number of distinct cod
e entities provided. In this report, we describe an algorithm that exe
cutes this strategy in the M-Technology environment.