Hypermedia and Randomized Algorithms for Medical Expert Systems

From Tetherless World Wiki

Jump to: navigation, search

Citation: R. Martin Chavez. (1989) Hypermedia and Randomized Algorithms for Medical Expert Systems. In KSL-89-14, 1989.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author R. Martin Chavez
title Hypermedia and Randomized Algorithms for Medical Expert Systems
number KSL-89-14
institution Knowledge Systems, AI Laboratory
address Washington D.C.
year 1989
Bibtex more
Access Paper
abstract Hypermedia and randomized algorithms for medical expert systems KNET is an environment for constructing probabilistic, knowledge-intensive systems within the axiomatic framework of decision theory. The KNET architecture defines a complete separation between the hypermedia user interface on the one hand, and the representation and management of expert opinion on the other. KNET offers a choice of algorithms for probabilistic inference. My coworkers and I have used KNET to build consultation systems for lymph-node pathology, bone-marrow transplantation therapy, clinical epidemiology, and alarm management in the intensive-care unit.Most important, KNET contains a fully polynomial randomized approximation scheme (fpras) for the difficult and almost certainly intractable problem of Bayesian inference. My algorithm can, in many circumstances, perform efficient approximate inference in large and richly interconnected models of medical diagnosis. In this article, I describe the architecture of KNET, construct a randomized algorithm for probabilistic inference, and analyze the algorithm's performance. Finally, I characterize my algorithm's empiric behavior and explore its potential for parallel speedups. From design to implementation, then, KNET clearly demonstrates the crucial interaction between theoretical computer science and medical informatics.

KSL Technical Report ID: KSL-89-14
Facts about Hypermedia and Randomized Algorithms for Medical Expert SystemsRDF feed
Abstract Hypermedia and randomized algorithms for m … Hypermedia and randomized algorithms for medical expert systems KNET is an environment for constructing probabilistic, knowledge-intensive systems within the axiomatic framework of decision theory. The KNET architecture defines a complete separation between the hypermedia user interface on the one hand, and the representation and management of expert opinion on the other. KNET offers a choice of algorithms for probabilistic inference. My coworkers and I have used KNET to build consultation systems for lymph-node pathology, bone-marrow transplantation therapy, clinical epidemiology, and alarm management in the intensive-care unit.Most important, KNET contains a fully polynomial randomized approximation scheme (fpras) for the difficult and almost certainly intractable problem of Bayesian inference. My algorithm can, in many circumstances, perform efficient approximate inference in large and richly interconnected models of medical diagnosis. In this article, I describe the architecture of KNET, construct a randomized algorithm for probabilistic inference, and analyze the algorithm's performance. Finally, I characterize my algorithm's empiric behavior and explore its potential for parallel speedups. From design to implementation, then, KNET clearly demonstrates the crucial interaction between theoretical computer science and medical informatics. computer science and medical informatics.
Address Washington D.C.  +
Author R. Martin Chavez  +
Bibtype techreport  +
Has author R. Martin Chavez  +
Has identifier KSL-89-14  +
Has publishing details 1989  +
Has title Hypermedia and Randomized Algorithms for Medical Expert Systems  +
Has where published KSL-89-14  +
Has year 1989  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-89-14  +
Number KSL-89-14  +
Process note YES  +
Title Hypermedia and Randomized Algorithms for Medical Expert Systems  +
Year 1989  +
Retrieved from "http://tw.rpi.edu/wiki/Hypermedia_and_Randomized_Algorithms_for_Medical_Expert_Systems"
Personal tools