By Topic

Using neural networks for modeling the input requirements of electronic medical record systems

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
S. E. Spenceley ; Sch. of Comput. & Inf. Sci., Univ. of South Australia, SA, Australia ; J. R. Warren

This paper presents work that has been conducted towards predicting user input requirements with view to making an intelligent interface to support data input in the context of an on-line electronic medical record system. The paper investigates how an artificial neural network, the self-organising feature map (SOM) suggested by Kohonen, may cluster patient data. Separate Bayesian probability models (for treatment given diagnoses) are derived for each cluster class (on a SOM with 25 output layer nodes). Clusterings are made on the basis of aggregate diagnoses over all the visits for particular patients considering longitudinal sequences of n>=5, n>=10, n>=15 and n >=20 visits. Clusterings produced from the longest visit sequence (n>=20) are found to be most useful in clustering the data. The predictions made from the 25 separate probability models compare most favorably with the predictions made by a single probability model derived from unclustered data. Clustering based upon shelter visit sequences also makes an improvement to the predictions made with unclustered data. The improved predictive power is explained by the use of longitudinal information, i.e. patient history. The improved predictive power translates into better prediction of the input requirements and hence user need.

Published in:

Systems Sciences, 1999. HICSS-32. Proceedings of the 32nd Annual Hawaii International Conference on  (Volume:Track6 )

Date of Conference:

5-8 Jan. 1999