MYCIN was a very early decision making system developed in the 1970s by Edward Shortliffe at Standard University. The software was programmed in LISP, which is similar to FORTRAN in age and programming technique. MYCIN was created in the same laboratory that brought about Dendral, an artificial intelligence system created in the 1960s.

The system was designed to identify bacteria which cause serious infections, such as bacteremia and meningitis, and to recommend a proper dosage of antibiotics based on the patient’s weight. Several antibiotics have the suffix, “-mycin” which is why the name was chosen.

Presently, electronic medical records or EMRs, have built-in decision support systems to aid a physician in selecting the right diagnosis for a specific patient. There is also electronic prescribing or e-prescribing, which has a built-in decision system as well. This system double checks the physician’s prescription and dosage levels. MYCIN was one of the first programs to do such a task.

Target Users:
Physicians, Medical Students, Paramedics

Edward Shortliffe (1972)
Department of Medicine and Computer Science
Heuristic Programming Project
Stanford University School of Medicine, California

Typical Consultation:
An example can be viewed at the University of Surrey. The application is text-based and asks several questions requiring input. The first three questions request a patient’s background information, including their first and last name, age and gender. In the example, the system automatically corrects a typing error of the word “Blood.”

There are then several questions to identify the significance of the infecting organism. After about 50-60 questions, MYCIN prints the diagnostic hypotheses on which the therapy and antibiotics will be based off of. MYCIN will then proceed through each rule, determining if the rule is false or true, and then will go onto the next rule. At the end of the consultation, all the answers will be compiled to discover which drugs should be selected.

Overall Goal:
  1. Create a patient context tree
  2. Decide if there is an organism that requires therapy
  3. Determine which drugs are potentially useful for the illness and select the best choice of antibiotic


Logical Layout of MYCIN

  1. University of Surrey.
  2. Wikipedia.
  3. Control Conditions in MYCIN: A Case Study, Colorado State University.