DocumentCode :
3309739
Title :
Artificial neural network-enabled prognostics for patient health management
Author :
Ghavami, P. ; Kapur, K.
Author_Institution :
Harborview Med. Center, UW Med., Seattle, WA, USA
fYear :
2012
fDate :
18-21 June 2012
Firstpage :
1
Lastpage :
8
Abstract :
Prognostics and prediction of patients´ short term physiological health status are of critical importance in medicine because they afford medical interventions that prevent escalating medical complications. This study proposes a prognostics engine to predict patient physiological status. The prognostics engine builds models from historical clinical data using neural network as its computational kernel. This study compared accuracy of various neural network models. Given the diversity of clinical data and disease conditions, no single model is ideal for all medical cases. Certain algorithms are more accurate than others depending on the type, amount and diversity of possible outcomes. Utilizing multiple neural network algorithms is a sound approach to building a generalizable prognostics engine. The study proposes using an ensemble of algorithms and an oracle, an overseer program to select the most accurate combination of the predictive models that is most suited for a particular disease prediction.
Keywords :
health care; medical computing; neural nets; patient diagnosis; artificial neural network-enabled prognostics; computational kernel; historical clinical data; medical cases; medical complications; medical interventions; neural network algorithms; neural network models; patient health management; patient physiological status; patient short term physiological health status prediction; patient short term physiological health status prognostics; predictive models; prognostics engine; Artificial neural networks; Biological system modeling; Biomedical monitoring; Diseases; Mathematical model; Medical diagnostic imaging; Predictive models; Neural networks; Prognostics; healthcare;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Prognostics and Health Management (PHM), 2012 IEEE Conference on
Conference_Location :
Denver, CO
Print_ISBN :
978-1-4673-0356-9
Type :
conf
DOI :
10.1109/ICPHM.2012.6299521
Filename :
6299521
Link To Document :
بازگشت