Ensemble learning for heartbeat classification using adaptive orthogonal transformations

Description

Recent advances in biomedical engineering make it possible to record physiological signals in various ways. For instance, blood pressure, electrocardiogram (ECG), electroencephalogram (EEG), electromyogram (EMG) can be measured by portable devices, which highly increase the demand for computer-assisted interpretation and analysis of these signals. In this work, we are focusing on the problem of heartbeat classification in ECGs. Following the recommendations of AAMI, the heartbeats should be classified into five categories: supraventricular (S), ventricular (V), fusion (F), unknown (Q), and normal (N). This is a complex task including preprocessing steps, feature extraction, training and evaluating machine learning algorithms. Our goal is to examine the potential of different adaptive signal models and combine them via ensemble learning in order to improve the individual classification results.

Period	21 Feb 2019
Event title	International Conference on Computer Aided Systems Theory (EUROCAST 2019)
Event type	Conference
Location	SpainShow on map