| Title | Event-based Neural Network for ECG Classification with Delta Encoding and Early Stopping |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Jobst, Matthias, Liu, Chen, Partzsch, Johannes, Yan, Yexin, Kappel, David, Gonzalez, Hector A., Ji, Yue, Vogginger, Bernhard, Mayr, Christian |
| Conference Name | 2020 6th International Conference on Event-Based Control, Communication, and Signal Processing (EBCCSP) |
| Keywords | Computer architecture, delta encoding, early stopping, ECG classification, Electrocardiography, encoding, event-based, feature extraction, Hardware, human factors, human in the loop, machine learning, pubcrawl, Recurrent neural networks, Task Analysis, Training |
| Abstract | We present a scalable architecture based on a trained filter bank for input pre-processing and a recurrent neural network (RNN) for the detection of atrial fibrillation in electrocardiogram (ECG) signals, with the focus on enabling a very efficient hardware implementation as application-specific integrated circuit (ASIC). Our already very efficient base architecture is further improved by replacing the RNN with a delta-encoded gated recurrent unit (GRU) and adding a confidence measure (CM) for terminating the computation as early as possible. With these optimizations, we demonstrate a reduction of the processing load of 58 % on an internal dataset while still achieving near state-of-the-art classification results on the Physionet ECG dataset with only 1202 parameters. |
| DOI | 10.1109/EBCCSP51266.2020.9291357 |
| Citation Key | jobst_event-based_2020 |
Event-based Neural Network for ECG Classification with Delta Encoding and Early Stopping