INTERPRETATION OF DOPPLER BLOOD-FLOW VELOCITY WAVE-FORMS USING NEURALNETWORKS

Authors
BAYKAL N REGGIA JA YALABIK N ERKMEN A BEKSAC MS
Citation
N. Baykal et al., INTERPRETATION OF DOPPLER BLOOD-FLOW VELOCITY WAVE-FORMS USING NEURALNETWORKS, Journal of the American Medical Informatics Association, 1994, pp. 865-869
Citations number
10
Categorie Soggetti
Information Science & Library Science","Medicine Miscellaneus","Computer Science Information Systems
Journal title
Journal of the American Medical Informatics Association → ACNP
ISSN journal
10675027
Year of publication
1994
Supplement
S
Pages
865 - 869
Database
ISI
SICI code
1067-5027(1994):<865:IODBVW>2.0.ZU;2-5
Abstract
Doppler umbilical artery blood flow velocity waveform measurement is u sed in perinatal surveillance for the evaluation of pregnancy status. There is an ongoing debate on the predictive value of doppler measurem ents concerning the critical effect of the selection of parameters for the evaluation of doppler output. In this paper, we describe how neur al network methods can be used both to discover relevant classificatio n features and subsequently to classify patients. Classification accur acy varied from 92-99% correct.