SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster\nAnalysis and Multilevel Noise Estimation

0487043 - ÚPT 2019 RIV FR eng C - Conference Paper (international conference)
Smíšek, Radovan - Hejč, J. - Ronzhina, M. - Němcová, A. - Maršánová, L. - Chmelík, J. - Kolářová, J. - Provazník, I. - Smítal, L. - Vítek, M.
SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster
Analysis and Multilevel Noise Estimation.
Computing in Cardiology 2017. Vol. 44. Rennes: Computing in Cardiology, 2017, s. 1-4. E-ISSN 2325-887X.
[Computing in Cardiology 2017. Rennes (FR), 24.09.2017-27.09.2017]
R&D Projects: GA ČR GAP102/12/2034
Institutional support: RVO:68081731
Keywords : ECG classifications * global feature * cross-validation technique
OECD category: Medical engineering
http://www.cinc.org/archives/2017/

Smartphone-based ECG devices comprise great potential in screening for arrhythmias. However, its feasibility is limited by poor signal quality leading to incorrect rhythm classification. In this study, advanced method for automatic classification of normal rhythm (N), atrial fibrillation (A), other rhythm (O), and noisy records (P) is introduced. Methods: Two-step SVM approach followed by simple threshold based rules was used for data classification. In the first step, various features were derived from separate beats to represent particular events (normal as well as pathological and artefacts) in more detail. Output of the first classifier was used to calculate global features describing entire ECG. These features were then used to train the second classification model. Both classifiers were evaluated on training set via cross-validation technique, and additionally on hidden testing set. Results: In the Phase II of challenge, total F1 score of the method is 0.81 and 0.84 within hidden challenge dataset and training set, respectively. Particular F1 scores within hidden challenge dataset are 0.90 (N), 0.81 (A), 0.72 (O), and 0.55 (P). Particular F1 scores within training set are 0.91 (N), 0.85 (A), 0.76 (O), and 0.73 (P).
Permanent Link: http://hdl.handle.net/11104/0284366