Title :
Detection of late potentials in electrocardiogram signals using artificial neural networks
Author :
Baykal, Ibrahim Cem ; Yilmaz, Atilla ; Kwan, H.K. ; Jullien, G.A.
Author_Institution :
Dept. of Electr. & Comput. Eng., Windsor Univ., Ont., Canada
Abstract :
Ventricular late potentials (LPs) are high-frequency low amplitude signals, which occur at the end of the QRS complex in electrocardiogram (ECG) signals. Though LPs are not valuable as a predictor of arrhythmic events and sudden cardiac death, there is a 95% probability of survival for a patient (after suffering damage in the myocardium), who does not have LPs in his/her ECG signal. An artificial neural network (ANN) model is used to detect LPs. The last 40 ms of the QRS segment is fed to the network along with the three time domain parameters, which are used as a standard way of predicting LPs. Training methods of ANN are discussed for this specific case
Keywords :
electrocardiography; learning (artificial intelligence); medical signal detection; medical signal processing; neural nets; ANN model; ANN training methods; ECG signals; HF low amplitude signals; QRS complex; artificial neural network model; electrocardiogram signals; late potentials detection; myocardium damage; time domain parameters; ventricular late potentials; Algorithm design and analysis; Artificial neural networks; Electrocardiography; Filters; Frequency; Intelligent networks; Signal analysis; Signal processing algorithms; Virtual manufacturing; Voltage;
Conference_Titel :
Circuits and Systems, 2000. Proceedings of the 43rd IEEE Midwest Symposium on
Conference_Location :
Lansing, MI
Print_ISBN :
0-7803-6475-9
DOI :
10.1109/MWSCAS.2000.951465
