Source:http://linkedlifedata.com/resource/pubmed/id/12575465
Switch to
| Predicate | Object |
|---|---|
| rdf:type | |
| lifeskim:mentions | |
| pubmed:dateCreated |
2003-2-10
|
| pubmed:abstractText |
The usefulness of a new way to optimize the cooperation of trained neural networks for automatic one-channel sleep stage analysis using genetic programming and performance evaluation by including the interrater reliability are the focus of our paper. The one-channel sleep classification could be significantly improved by the optimization. The software tool HENNE, with its genetic programming compartment was developed for this purpose. The tool has proved to be useful for searching for optima in difficult goal surfaces. To contribute to the general discussion about the benefit of the automatic one-channel sleep analysis on the basis of the frontal site, we tried to evaluate our results before the background of the interrater variability. Comparing the kappa statistics of different independent studies with our results, we concluded that there are no dramatic differences as a rule and that QUISI is a useful device as a presleep laboratory and ambulatory diagnostic tool.
|
| pubmed:language |
eng
|
| pubmed:journal | |
| pubmed:citationSubset |
IM
|
| pubmed:status |
MEDLINE
|
| pubmed:issn |
0379-0355
|
| pubmed:author | |
| pubmed:issnType |
Print
|
| pubmed:volume |
24 Suppl D
|
| pubmed:owner |
NLM
|
| pubmed:authorsComplete |
Y
|
| pubmed:pagination |
27-32
|
| pubmed:dateRevised |
2004-11-17
|
| pubmed:meshHeading | |
| pubmed:year |
2002
|
| pubmed:articleTitle |
Genetic programming approach for the optimal selection of combinations of neuronal networks to classify sleep stages by QUISI.
|
| pubmed:affiliation |
Dept. of Electrical Engineering, University of Applied Science of Schmalkalden, Germany. bschmitt@e-technik.fh-schmalkalden.de
|
| pubmed:publicationType |
Journal Article
|