MYCIN was an early expert system developed in the 1970s to diagnose infectious diseases. It used a knowledge base of rules provided by experts and an inference engine to identify bacteria causing infections, determine appropriate antibiotics, and provide treatments. MYCIN could also explain its reasoning. It demonstrated that expert systems could perform competently in domains requiring specialized knowledge.
2. OUTLINE
HISTORY
MYCIN : The Problem
System Components
Expert System Structure
STAGES: in Diagnosis and Treatment
MYCIN Architecture
Explanation
Summary
3. HISTORY
MYCIN was developed at Stanford
University in the mid 1970s.
Project spans a decade
Research started in 1972.
Original Implementation
completed in 1976.
Research continued into the
1980s.
4. MYCIN : The Problem
Roberts & Visconti [1972]:
Only 13% of patients are treated.
66% are being irrational treatment.
21% are being questionable treatment.
• Irrationality means, for example :
Using a contraindicated combination.
Using the wrong agent for a specific
organism.
Not taking the required cultures.
5. Design Parameter
Program must be competent and easy
to use.
Must handle a large changing body of
knowledge.
Interact with human users.
Must take time into account.
Work with incomplete and uncertain
information.
6. SYSTEM COMPONENTS
• Consultation System
ask question.
Draws conclusions.
Given advice .
• Explanation System
Translation ruler
• Rule Acquisition / Modification System
8. Stages in Diagnosis and
Treatment
Decide if there is any significant infection.
Identify the causing organism by clinical and
laboratory evidence.
Decide what antibiotics agent the organism are
sensitive too.
Prescribe the optimal drug combinations for the
particular care.
10. Consultation System
Consultation System Has
Performs Diagnosis and Therapy
Selection
Control Structure reads Static DB (rules)
and read/writes
to Dynamic DB
(patient/ context).
Linked to Explanation
Terminal interface to Physician.
11. Explanation System
Provides reasoning why a conclusion has been
made , or
why a question is being asked.
• Record rule invocation and associates them with
questions and rules invoked.
• Use rule index to retrieve particular rules in
answer to question .
• Why and how questions are answered using goal
tree.
12. Knowledge Acquisition Program
Extend static DB via Dialogue with experts
Dialogue driven by Experts
Requires minimal training for Experts.
Allows for Incremental Competence , NOT an
ALL –or- Nothing model.
IF-THEN symbolic logic was found to be easy for
Experts to learn, and required little training by the
MYCIN training.
13. SUMMARY
MYCIN combines the advantages of general rule
based system with the advantages of an “inexact”
reasoning.
• MYCIN has not addressed how to convert from
human terms to certainties how to normalize
across different people’s
how far to propagate certainty factor changes
based on new evidences how to provide
feedback to database to improve
certainty factor accuracy.