ECG Classification with a Convolutional Recurrent Neural Network

09/28/2020

∙

We developed a convolutional recurrent neural network to classify 12-lead ECG signals for the challenge of PhysioNet/ Computing in Cardiology 2020 as team Pink Irish Hat. The model combines convolutional and recurrent layers, takes sliding windows of ECG signals as input and yields the probability of each class as output. The convolutional part extracts features from each sliding window. The bi-directional gated recurrent unit (GRU) layer and an attention layer aggregate these features from all windows into a single feature vector. Finally, a dense layer outputs class probabilities. The final decision is made using test time augmentation (TTA) and an optimized decision threshold. Several hyperparameters of our architecture were optimized, the most important of which turned out to be the choice of optimizer and the number of filters per convolutional layer. Our network achieved a challenge score of 0.511 on the hidden validation set and 0.167 on the full hidden test set, ranking us 23rd out of 41 in the official ranking.

READ FULL TEXT

ECG Classification with a Convolutional Recurrent Neural Network

Classification of 12-Lead ECG Signals with Bi-directional LSTM Network

Cardiac Arrhythmia Detection from ECG with Convolutional Recurrent Neural Networks

Method to Annotate Arrhythmias by Deep Network

Arrhythmia Classification from the Abductive Interpretation of Short Single-Lead ECG Records

MSW-Transformer: Multi-Scale Shifted Windows Transformer Networks for 12-Lead ECG Classification

Combining Scatter Transform and Deep Neural Networks for Multilabel Electrocardiogram Signal Classification

Detection of Obstructive Sleep Apnoea Using Features Extracted from Segmented Time-Series ECG Signals Using a One Dimensional Convolutional Neural Network

ECG Classification with a Convolutional Recurrent Neural Network

Related Research

Classification of 12-Lead ECG Signals with Bi-directional LSTM Network

Cardiac Arrhythmia Detection from ECG with Convolutional Recurrent Neural Networks

Method to Annotate Arrhythmias by Deep Network

Arrhythmia Classification from the Abductive Interpretation of Short Single-Lead ECG Records

MSW-Transformer: Multi-Scale Shifted Windows Transformer Networks for 12-Lead ECG Classification

Combining Scatter Transform and Deep Neural Networks for Multilabel Electrocardiogram Signal Classification

Detection of Obstructive Sleep Apnoea Using Features Extracted from Segmented Time-Series ECG Signals Using a One Dimensional Convolutional Neural Network