|
|
Clustering of ECG cycles
Němečková, Karolína ; Kozumplík, Jiří (referee) ; Ronzhina, Marina (advisor)
This bachelor thesis deals with application of cluster analysis to different ECG records in order to identify particular cardiac pathologies. The work is mainly focused on the detection of premature atrial and premature ventricular beats. Presented approach is based on the signal correlation and further beat type identification and beats clustering via specific ECG features. By evaluation the method on test data, we obtained TPR 73.40 %, FPR 91.00 %, PPV 29.00 %, ACC 90.00 %, F1 41.40 % for PAC detection and TPR 76.50 %, FPR 94.20 %, PPV 45.90 %, ACC 93.10 %, F1 57.40 % for PVC detection. Pure F1 and PPV is due to high number of false positive detections mainly in noisy ECG or ECG with manifested atrial fibrillation.
|
|
|
Clustering of ECG cycles
Němečková, Karolína ; Kozumplík, Jiří (referee) ; Ronzhina, Marina (advisor)
This bachelor thesis deals with application of cluster analysis to different ECG records in order to identify particular cardiac pathologies. The work is mainly focused on the detection of premature atrial and premature ventricular beats. Presented approach is based on the signal correlation and further beat type identification and beats clustering via specific ECG features. By evaluation the method on test data, we obtained TPR 73.40 %, FPR 91.00 %, PPV 29.00 %, ACC 90.00 %, F1 41.40 % for PAC detection and TPR 76.50 %, FPR 94.20 %, PPV 45.90 %, ACC 93.10 %, F1 57.40 % for PVC detection. Pure F1 and PPV is due to high number of false positive detections mainly in noisy ECG or ECG with manifested atrial fibrillation.
|
guest ::
