Could a computer diagnose blood infections better than most doctors in 1978?

The answer

MYCIN, developed at Stanford University by Edward Shortliffe between 1972 and 1976, was an expert system designed to identify bacteria causing severe infections and recommend antibiotics. It used approximately 600 rules derived from expert physicians and employed certainty factors to handle uncertain medical data. In a landmark 1979 evaluation, MYCIN prescribed acceptable therapy in about 69% of cases, outperforming the 42.5% to 62.5% rate of five Stanford faculty members. Despite its impressive performance, MYCIN was never deployed clinically due to ethical and legal concerns about allowing a computer to make medical treatment decisions, as well as the practical difficulty of entering patient data manually.