Title :
Estimating ventilation using artificial neural networks in intensive care units
Author :
Tong, Yanling ; Frize, Monique ; Walker, Robin
Author_Institution :
S.I.T.E., Ottawa Univ., Ont., Canada
Abstract :
Few medical researchers have achieved correct classification rates (CCR) with artificial neural networks (ANNs). Trigg et al. [see Procs. of the World Congress on Biomed. Engng. and Phys., Nice, France, Sept. 1997] obtained good results using an adult intensive care unit (AICU) database. This paper extends Trigg´s technique-ANNs with weight-elimination, to estimate ventilation with a Neonatal Intensive Care Unit (NICU) database. Encouraging results were obtained in terms of Correct Classification Rate (CCR) and Average Squared Error (ASE) and the adult model could be successfully applied to neonatal patients
Keywords :
biomedical measurement; medical computing; neural nets; paediatrics; patient care; pneumodynamics; adult intensive care unit database; adult model; artificial neural networks; average squared error; correct classification rate; neonatal intensive care unit database; ventilation estimation; Artificial neural networks; Databases; Environmental economics; Hospitals; Intelligent networks; Oxygen; Pediatrics; Testing; Ventilation; Weight control;
Conference_Titel :
[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-5674-8
DOI :
10.1109/IEMBS.1999.804395
