Long-Time Prediction of Arrhythmic Cardiac Action Potentials Using Recurrent Neural Networks and Reservoir Computing

RT @GTCSE: Check out this latest paper from CSE researchers showing #ML time-series prediction methods can forecast synthetic & experimenta…

Check out this latest paper from CSE researchers showing #ML time-series prediction methods can forecast synthetic & experimental cardiac action potentials better than other methods!🫀https://t.co/Abc9ndkkRe

Our paper is finally published and is available online. In this paper, we showed how various architectures of Echo State Networks (ESNs) can accurately predict nonlinear time series in real-world applications, while they’re about 100X faster than other RNN

🎉Congrats to CSE Associate Prof. Elizabeth Cherry, Ph.D. student Shahrokh Shahi, & Postdoc Christopher D. Marcotte! 🎉 Their paper on predicting cardiac voltage data using RNNs and ESNs is now published in @FrontPhysiol and can be read here: https://t.