Collecting and Interpreting Temporal Data in an Expert System
Abstract
Heart Disease Program
State of example case to be diagnosed
Temporal properties of data
Temporal patterns needed
Physician enters case:
Physician adds details
Physician describes chest pain
Temporal menus needed to capture patient state
PPT Slide
Intervals determined by HDP
Challenges interpreting descriptions
Temporal boundary knowledge
Findings indicating a progressive disease appear in several ways
Defaults are needed to fill in the data the user leaves unspecified
Defaults can lead to mistakes
HDP diagnosis
Future directions
References
Email: hamish@mit.edu
Home Page: http://medg.lcs.mit.edu/people/hamish/index~3.htm
Other information: Lecture 1