English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper proposes piece-wise matching layer as a novel layer inrepresentation learning methods for electrocardiogram (ECG) classification.Despite the remarkable performance of representation learning methods in theanalysis of time series, there are still several challenges associated withthese methods ranging from the complex structures of methods, the lack ofgenerality of solutions, the need for expert knowledge, and large-scaletraining datasets. We introduce the piece-wise matching layer that works basedon two levels to address some of the aforementioned challenges. At the firstlevel, a set of morphological, statistical, and frequency features andcomparative forms of them are computed based on each periodic part and itsneighbors. At the second level, these features are modified by predefinedtransformation functions based on a receptive field scenario. Several scenariosof offline processing, incremental processing, fixed sliding receptive field,and event-based triggering receptive field can be implemented based on thechoice of length and mechanism of indicating the receptive field. We proposedynamic time wrapping as a mechanism that indicates a receptive field based onevent triggering tactics. To evaluate the performance of this method in timeseries analysis, we applied the proposed layer in two publicly availabledatasets of PhysioNet competitions in 2015 and 2017 where the input data is ECGsignal. We compared the performance of our method against a variety of knowntuned methods from expert knowledge, machine learning, deep learning methods,and the combination of them. The proposed approach improves the state of theart in two known completions 2015 and 2017 around 4% and 7% correspondinglywhile it does not rely on in advance knowledge of the classes or the possibleplaces of arrhythmia.

Piece-wise Matching Layer in Representation Learning for ECG Classification