A machine-learning approach for long-term prediction of experimental cardiac action potential time series using an autoencoder and echo state networks
- 1 June 2022
- journal article
- research article
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 32 (6), 063117
- https://doi.org/10.1063/5.0087812
Abstract
Computational modeling and experimental/clinical prediction of the complex signals during cardiac arrhythmias have the potential to lead to new approaches for prevention and treatment. Machine-learning (ML) and deep-learning approaches can be used for time-series forecasting and have recently been applied to cardiac electrophysiology. While the high spatiotemporal nonlinearity of cardiac electrical dynamics has hindered application of these approaches, the fact that cardiac voltage time series are not random suggests that reliable and efficient ML methods have the potential to predict future action potentials. This work introduces and evaluates an integrated architecture in which a long short-term memory autoencoder (AE) is integrated into the echo state network (ESN) framework. In this approach, the AE learns a compressed representation of the input nonlinear time series. Then, the trained encoder serves as a feature-extraction component, feeding the learned features into the recurrent ESN reservoir. The proposed AE-ESN approach is evaluated using synthetic and experimental voltage time series from cardiac cells, which exhibit nonlinear and chaotic behavior. Compared to the baseline and physics-informed ESN approaches, the AE-ESN yields mean absolute errors in predicted voltage 6–14 times smaller when forecasting approximately 20 future action potentials for the datasets considered. The AE-ESN also demonstrates less sensitivity to algorithmic parameter settings. Furthermore, the representation provided by the feature-extraction component removes the requirement in previous work for explicitly introducing external stimulus currents, which may not be easily extracted from real-world datasets, as additional time series, thereby making the AE-ESN easier to apply to clinical data.Keywords
Funding Information
- National Science Foundation (CMMI-2011280)
- National Science Foundation (CMMI-1762553)
- National Heart, Lung, and Blood Institute (1R01HL143450)
This publication has 64 references indexed in Scilit:
- Effects of Pacing Site and Stimulation History on Alternans Dynamics and the Development of Complex Spatiotemporal Patterns in Cardiac TissueFrontiers in Physiology, 2013
- Mechanisms of ventricular arrhythmias: a dynamical systems-based perspectiveAmerican Journal of Physiology-Heart and Circulatory Physiology, 2012
- Control of electrical alternans in simulations of paced myocardium using extended time-delay autosynchronizationPhysical Review E, 2007
- Spatially discordant voltage alternans cause wavebreaks in ventricular fibrillationHeart Rhythm, 2007
- Nonlinear dynamics of heart rhythm disordersPhysics Today, 2007
- A Fast Learning Algorithm for Deep Belief NetsNeural Computation, 2006
- Suppression of alternans and conduction blocks despite steep APD restitution: electrotonic, memory, and conduction velocity restitution effectsAmerican Journal of Physiology-Heart and Circulatory Physiology, 2004
- Continuous Representations of Time-Series Gene Expression DataJournal of Computational Biology, 2003
- Spatiotemporal Chaos in a Simulated Ring of Cardiac CellsPhysical Review Letters, 1997
- Electrical alternans and spiral wave breakup in cardiac tissueChaos: An Interdisciplinary Journal of Nonlinear Science, 1994