TY - GEN
T1 - A switching feature extraction system for ECG heartbeat classification
AU - De Chazal, Philip
PY - 2013
Y1 - 2013
N2 - This study compared two methods for extracting ECG waveshape features useful for heartbeat classification. The first method (segmented waveshape features) sampled the ECG waveshape between the P and T waveform boundaries, calculated QRS and T wave durations, and used RR-interval features. The second method (fixed interval waveshape features) used a fixed window to capture the ECG waveshape and RR-interval features. Data were obtained from the MIT-EIH arrhythmia database. We investigated the problem of discriminating between normal, supraventricular (SVEE), ventricular (VEE), fusion and unknown beat classes. When the P, QRS and T wave boundaries could be found reliably, the segmented waveshape features resulted in more balanced performance for discriminating SVEE and VEE beats than the fixed interval waveshape features. A hybrid approach using segmented and fixed interval features when waveform boundaries could be reliably found, and fixed interval features otherwise was the most robust solution. Using the AAMI recommendations for cardiac rhythm disturbances the hybrid approach resulted in a sensitivity of 69%, a positive predictivity of 31% and a false positive ratio (FPR) of6.6%for SVEE class. For the VEE class the sensitivity was 80%, the positivity predictivity was 85% and the FPR was 1.0%.
AB - This study compared two methods for extracting ECG waveshape features useful for heartbeat classification. The first method (segmented waveshape features) sampled the ECG waveshape between the P and T waveform boundaries, calculated QRS and T wave durations, and used RR-interval features. The second method (fixed interval waveshape features) used a fixed window to capture the ECG waveshape and RR-interval features. Data were obtained from the MIT-EIH arrhythmia database. We investigated the problem of discriminating between normal, supraventricular (SVEE), ventricular (VEE), fusion and unknown beat classes. When the P, QRS and T wave boundaries could be found reliably, the segmented waveshape features resulted in more balanced performance for discriminating SVEE and VEE beats than the fixed interval waveshape features. A hybrid approach using segmented and fixed interval features when waveform boundaries could be reliably found, and fixed interval features otherwise was the most robust solution. Using the AAMI recommendations for cardiac rhythm disturbances the hybrid approach resulted in a sensitivity of 69%, a positive predictivity of 31% and a false positive ratio (FPR) of6.6%for SVEE class. For the VEE class the sensitivity was 80%, the positivity predictivity was 85% and the FPR was 1.0%.
UR - http://handle.uws.edu.au:8081/1959.7/537299
M3 - Conference Paper
SN - 9781479908844
SP - 955
EP - 958
BT - Computing in Cardiology 2013. Vol. 40: September 22-25, 2013, Zaragoza, Spain
PB - Computing in Cardiology/IEEE
T2 - Computing in Cardiology
Y2 - 22 September 2013
ER -