A problem-solving model for protocol-based care: from e-oncocin to eon
From Semantic Portal Wiki
{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}
| Edit |
Reference: {{#vardefine:pagename|a problem-solving model for protocol-based care: from e-oncocin to eon }}
- [[]]
bibtex
{{#vardefine:pagename|A problem-solving model for protocol-based care: from e-oncocin to eon }}{{#vardefine:key| }}
abstract: Decision-support programs can plan medical therapy using a problem-solving method called episodic skeletal-plan refinement (ESPR). The e-ONCOCIN system incorporates a fixed form of the ESPR method that makes the program unable to reason about protocol-based care when treatment decisions must be made on the basis of abstractions of primary data, and when individual therapy actions may themselves require initiation of complex procedures. We have developed a new reasoning framework, called EON, in which problem-solving methods such as ESPR can be divided into discrete subtasks. EON invokes alternative, smaller-grained problem-solving methods for each subtask of a method. Subtasks of the smaller-grained methods may be solved by yet other methods, in a recursive manner. A formal model of the data processed by the problem-solving methods in EON simplifies knowledge acquisition and insulates developers from the details of how knowledge bases are implemented. The EON architecture provides considerable flexibility in constructing problem solvers, and offers system builders the ability to author knowledge bases that facilitate protocol-based care in a wide range of clinical areas.
download:
- paper:
- slides:
| Abstract | Decision-support programs can plan medical … Decision-support programs can plan medical therapy using a problem-solving method called episodic skeletal-plan refinement (ESPR). The e-ONCOCIN system incorporates a fixed form of the ESPR method that makes the program unable to reason about protocol-based care when treatment decisions must be made on the basis of abstractions of primary data, and when individual therapy actions may themselves require initiation of complex procedures. We have developed a new reasoning framework, called EON, in which problem-solving methods such as ESPR can be divided into discrete subtasks. EON invokes alternative, smaller-grained problem-solving methods for each subtask of a method. Subtasks of the smaller-grained methods may be solved by yet other methods, in a recursive manner. A formal model of the data processed by the problem-solving methods in EON simplifies knowledge acquisition and insulates developers from the details of how knowledge bases are implemented. The EON architecture provides considerable flexibility in constructing problem solvers, and offers system builders the ability to author knowledge bases that facilitate protocol-based care in a wide range of clinical areas. ed care in a wide range of clinical areas. |
| Address | Palexpo Geneva, Switzerland + |
| Author | Mark A. Musen +, Samson W. Tu +, and Yuval Shahar + |
| Bibtype | techreport + |
| Institution | Knowledge Systems, AI Laboratory + |
| Key | KSL-91-61 + |
| Number | KSL-91-61 + |
| Tag | Computer science + |
| Title | A Problem-Solving Model for Protocol-Based Care: From e-ONCOCIN to EON + |
| Tr id | KSL-91-61 + |
| Year | 1992 + |
