Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart disease

From Tetherless World Wiki

Jump to: navigation, search

Citation: Lucila Ohno-Machado and Mark A. Musen. (1996) Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart disease. In KSL-96-10, February,1996.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author Lucila Ohno-Machado and Mark A. Musen
title Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart disease
number KSL-96-10
institution Knowledge Systems, AI Laboratory
year 1996
month February
Bibtex more
note Medical Computer Science
Access Paper
abstract Medical researchers who perform prognostic modeling usually oversimplify the problem by choosing a single point in time to predict outcomes (e.g., death in five years). This approach not only fails to differentiate patterns of disease progression, but also wastes important information that is usually available in time-oriented research data bases. The adequate use of time-oriented data bases can improve the performance of prognostic systems if the interdependencies among prognoses at different intervals of time are explicitly modeled. In such models, predictions for a certain interval of time (e.g., death within one year) are influenced by predictions made for other intervals, and prognostic survival curves that provide consistent estimates for several points in time can be produced. We developed a system of neural network models that makes use of time- oriented data to predict development of coronary heart disease (CHD), using a set of 2594 patients. The output of the neural network system was a prognostic curve representing survival without CHD, and the inputs were the values of demographic, clinical, and laboratory variables. The system of neural networks was trained by backprogation and its results were evaluated in test sets of previously unseen cases. We showed that, by explicitly modeling time in the neural net work architecture, the performance of the prognostic index, measured by the area under the receiver operating characteristic (ROC) curve, was significantly improved (p<0.05).

KSL Technical Report ID: KSL-96-10
Facts about Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart diseaseRDF feed
Abstract Medical researchers who perform prognostic … Medical researchers who perform prognostic modeling usually oversimplify the problem by choosing a single point in time to predict outcomes (e.g., death in five years). This approach not only fails to differentiate patterns of disease progression, but also wastes important information that is usually available in time-oriented research data bases. The adequate use of time-oriented data bases can improve the performance of prognostic systems if the interdependencies among prognoses at different intervals of time are explicitly modeled. In such models, predictions for a certain interval of time (e.g., death within one year) are influenced by predictions made for other intervals, and prognostic survival curves that provide consistent estimates for several points in time can be produced. We developed a system of neural network models that makes use of time- oriented data to predict development of coronary heart disease (CHD), using a set of 2594 patients. The output of the neural network system was a prognostic curve representing survival without CHD, and the inputs were the values of demographic, clinical, and laboratory variables. The system of neural networks was trained by backprogation and its results were evaluated in test sets of previously unseen cases. We showed that, by explicitly modeling time in the neural net work architecture, the performance of the prognostic index, measured by the area under the receiver operating characteristic (ROC) curve, was significantly improved (p<0.05). , was significantly improved (p<0.05).
Author Lucila Ohno-Machado and Mark A. Musen  +
Bibtype techreport  +
Has author Lucila Ohno-Machado and Mark A. Musen  +
Has identifier KSL-96-10  +
Has publishing details February,1996  +
Has title Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart disease  +
Has where published KSL-96-10  +
Has year 1996  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-96-10  +
Month February  +
Note Medical Computer Science
Number KSL-96-10  +
Process note NO  +
Title Sequential versus standard neural networks for temporal pattern recognition: An example using the domain of coronary heart disease  +
Year 1996  +
Retrieved from "http://tw.rpi.edu/wiki/Sequential_versus_standard_neural_networks_for_temporal_pattern_recognition:_An_example_using_the_domain_of_coronary_heart_disease"
Personal tools