| Publication Type |
dissertation |
| School or College |
School of Medicine |
| Department |
Biomedical Informatics |
| Author |
Guo, Di |
| Title |
Development of the new best information algorithm for a medical expert system (Iliad) |
| Date |
1993-08 |
| Description |
Iliad is a diagnostic expert system for internal medicine. One important feature that Iliad offers is the ability to analyze a particular patient case and to determine the most cost-effective findings to pursue next at any stage of a work-up. The best information"" algorithm combines an information content calculation together with a cost factor. The calculations then provide a rank-ordering of the alternative patient findings according to cost-effectiveness. This dissertation presents a three-part study to evaluate the performance of different best information algorithms. In the first two parts of the study the suggestions about the next best data elements to pursue from different algorithms were collected for different vignettes. The performance of different algorithms was compared based on the judgments provided by expert clinicians. The results indicated that the current Iliad information content model could be improved by using a version of Shannon information content model. The third part of the study evaluated different best information algorithms by a simulation approach. The results indicated that two types of diagnostic behaviors could be simulated. The first type of behavior was characterized by pursuing more history and physical examination findings, less laboratory tests, less expensive work-ups, and more steps to solve a patient case. The second type of behavior was characterized by pursuing less history and physical examination findings, more laboratory tests, more expensive work-ups, and less steps to solve a patient case. The Shannon information content model accomplished work-ups that were significantly less costly than work-ups performed by the current LR (likelihood ratio) information content model. However, the Shannon model required additional computational resources and more history and physical examination steps than the LR model. Decisions regarding the implementation of alternative models require a balance of the relative merits of cost, steps, expert preference, and other important factors."" |
| Type |
Text |
| Publisher |
University of Utah |
| Subject |
Medical Informatics |
| Subject MESH |
Expert Systems; Diagnosis, Computer-Assisted; Knowledge Bases; Algorithms; Computer Simulation; Medical Informatics Applications; Diagnosis, Differential; Information Theory; Benchmarking; Probability Theory; Bayes Theorem |
| Dissertation Institution |
University of Utah |
| Dissertation Name |
Doctor of Philosophy |
| Language |
eng |
| Relation is Version of |
Digital reproduction of "Development of the new best information algorithm for a medical expert system (Iliad)." Spencer S. Eccles Health Sciences Library. |
| Rights Management |
Copyright © 1993 Di Guo |
| Format Medium |
application/pdf |
| Format Extent |
4,601,436 bytes |
| Identifier |
undthes,4355 |
| Source |
Original: University of Utah Spencer S. Eccles Health Sciences Library (no longer available). |
| Funding/Fellowship |
National Library of Medicine NLM Grants I01-LM-04604 and IR01-LM 052020. and John Morgan Fellowship. |
| Master File Extent |
4,601,503 bytes |
| ARK |
ark:/87278/s6z89f7w |
| Setname |
ir_etd |
| ID |
191389 |
| Reference URL |
https://collections.lib.utah.edu/ark:/87278/s6z89f7w |
