Noise-Aware Self-Supervised Denoising for Electrocardiogram Anomaly Detection
SSRN, ISSN: 1556-5068
2023
- 229Usage
Metric Options: Counts1 Year3 YearSelecting the 1-year or 3-year option will change the metrics count to percentiles, illustrating how an article or review compares to other articles or reviews within the selected time period in the same journal. Selecting the 1-year option compares the metrics against other articles/reviews that were also published in the same calendar year. Selecting the 3-year option compares the metrics against other articles/reviews that were also published in the same calendar year plus the two years prior.
Example: if you select the 1-year option for an article published in 2019 and a metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019. If you select the 3-year option for the same article published in 2019 and the metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019, 2018 and 2017.
Citation Benchmarking is provided by Scopus and SciVal and is different from the metrics context provided by PlumX Metrics.
Example: if you select the 1-year option for an article published in 2019 and a metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019. If you select the 3-year option for the same article published in 2019 and the metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019, 2018 and 2017.
Citation Benchmarking is provided by Scopus and SciVal and is different from the metrics context provided by PlumX Metrics.
Article Description
The electrocardiogram (ECG) is a non-invasive and straightforward diagnostic tool commonly employed to identify heart conditions. ECG anomaly detection aims to find abnormal patterns including disease data and invalid noise data. In this paper, we propose a simple and effective self-supervised anomaly detection technique called Noise Distribution Aware Autoencoder (NDAAE), where a novel proxy task is designed for self-supervised learning of normal ECG representations, by conducting noise-aware denoising of ECG signal, where the learned representation is robust enough to distinguish data with anomalous patterns from normal ones. Specifically, we employ the following procedure: Firstly, we randomly inject five distinct noise signals into the original ECG data. Subsequently, the ECG data with added noise is passed through an autoencoder to reconstruct the original ECG data without noise. Finally, we integrate the noise classification task, which guides the model to reduce noise effectively, into the auto-encoder framework. This leads to the acquisition of a more effective ECG representation, facilitating advanced anomaly detection capabilities. We evaluate our proposed method on real ECG datasets, and the empirical results illustrate that our model outperforms existing approaches in detecting anomalies.
Bibliographic Details
Provide Feedback
Have ideas for a new metric? Would you like to see something else here?Let us know