Title :
HMM based sensor fault detection in neuromuscular blockade control
Author :
Lemos, Joao M. ; Magalhaes, Hugo ; Mendonca, Teresa
Author_Institution :
INESC-ID/IST, Lisbon, Portugal
Abstract :
The problem of embedding sensor fault tolerance in the control of neuromuscular blockade of patients subject to anesthesia for surgery is considered. A hidden Markov model whose state, assumed unavailable for direct observation, is estimated using a nonlinear filtering algorithm (special case of Baum´s algorithm), is proposed for detecting the faults. The resulting algorithm is illustrated by simulations with realistic models of neuromuscular blockade.
Keywords :
electromyography; fault diagnosis; fault tolerant control; filtering theory; hidden Markov models; medical control systems; medical signal processing; muscle; neurophysiology; nonlinear filters; sensors; state estimation; surgery; Baum algorithm; HMM based sensor fault detection; anesthesia; evoked EMG; fault detection; fault tolerant control; hidden Markov model; nonlinear filtering algorithm; patients neuromuscular blockade control; sensor fault tolerance embedding problem; state estimation; surgery; Anesthesia; Biomedical measurement; Hidden Markov models; Markov processes; Neuromuscular; Surgery; Anaesthesia; Biomedicine; Control Applications; Fault Tolerant Control; HMM; Neuromuscular Blockade;
Conference_Titel :
Control Conference (ECC), 2007 European
Conference_Location :
Kos
Print_ISBN :
978-3-9524173-8-6
