Key result
TCN-AE outperforms state-of-the-art algorithms for unsupervised ECG anomaly detection in cardiac arrhythmias.
Why the study?
Learning temporal patterns in time series for anomaly detection is challenging, especially for periodic or quasiperiodic signals with complex temporal patterns.
TCN-AE, an unsupervised temporal convolutional network autoencoder, significantly improves anomaly detection in ECG time series compared to existing state-of-the-art algorithms.
Learning temporal patterns in time series remains a challenging task up until today. Particularly for anomaly detection in time series, it is essential to learn the underlying structure of a system’s normal behavior. Periodic or quasiperiodic signals with complex temporal patterns make the problem even more challenging: Anomalies may be a hard-to-detect deviation from the normal recurring pattern. In this paper, we present TCN-AE, a t emporal c onvolutional n etwork a uto e ncoder based on dilated convolutions. Contrary to many other anomaly detection algorithms, TCN-AE is trained in an unsupervised manner. The algorithm demonstrates its efficacy on a comprehensive real-world anomaly benchmark comprising electrocardiogram (ECG) recordings of patients with cardiac arrhythmia . TCN-AE significantly outperforms several other unsupervised state-of-the-art anomaly detection algorithms. Moreover, we investigate the contribution of the individual enhancements and show that each new ingredient improves the overall performance on the investigated benchmark.
No takes yet. Share an insight, caveat, or question.
Thill et al. (2021) studied Cardiac arrhythmia. TCN-AE (temporal convolutional network autoencoder) vs. Other unsupervised state-of-the-art anomaly detection algorithms was evaluated on Anomaly detection performance. TCN-AE, an unsupervised temporal convolutional network autoencoder, significantly outperformed several other state-of-the-art algorithms for anomaly detection in ECG recordings of cardiac arrhythmia.