ECG Heartbeat Classification: A Deep Transferable Representation

By Mohammad Kachuee, Shayan Fazeli, Majid Sarrafzadeh. 2018

518 citations recorded by the source corpus.

Electrocardiogram (ECG) can be reliably used as a measure to monitor the functionality of the cardiovascular system. Recently, there has been a great attention towards accurate categorization of heartbeats. While there are many commonalities between different ECG conditions, the focus of most studies has been classifying a set of conditions on a dataset annotated for that task rather than learning and employing a transferable knowledge between different tasks. In this paper, we propose a method based on deep convolutional neural networks for the classification of heartbeats which is able to accurately classify five different arrhythmias in accordance with the AAMI EC57 standard. Furthermore, we suggest a method for transferring the knowledge acquired on this task to the myocardial infarction (MI) classification task. We evaluated the proposed method on PhysionNet's MIT-BIH and PTB Diagnostics datasets. According to the results, the suggested method is able to make predictions with the average accuracies of 93.4% and 95.9% on arrhythmia classification and MI classification, respectively.

Subfields: Cardiology and Cardiovascular Medicine, Cognitive Neuroscience, Pulmonary and Respiratory Medicine

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cardiology and Cardiovascular Medicine

1. Matin Hashemi
2. Tomás Ward
3. Anne Marie Delaney
4. Mohammadhosein Oveisi
5. Eoin Brophy
6. Saeed Saadatnejad

In Cognitive Neuroscience

1. Matin Hashemi
2. Tomás Ward
3. Anne Marie Delaney
4. Mohammadhosein Oveisi
5. Eoin Brophy
6. Saeed Saadatnejad

In Pulmonary and Respiratory Medicine

1. Matin Hashemi
2. Tomás Ward
3. Anne Marie Delaney
4. Mohammadhosein Oveisi
5. Eoin Brophy
6. Saeed Saadatnejad

Citations in the corpus

• Interpretable and Steerable Sequence Learning via Prototypes (2019, methodology): “We can see that ProSeNet has comparable performance to LSTM, and even slightly better accuracy than Residual CNN [16].”
• Synthesis of Realistic ECG using Generative Adversarial Networks (2019, methodology): “The techniques listed below were applied to the original signals to clean the data in preparation for machine learning tasks [44].”
• LSTM-Based ECG Classification for Continuous Monitoring on Personal Wearable Devices (2018, methodology): “[21] proposed a deep CNN algorithm with 11 layers which is less accurate and about 100X slower compared to the proposed method.”
• Artificial Intelligence (AI) and Internet of Medical Things (IoMT) Assisted Biomedical Systems for Intelligent Healthcare (2022, background): “proposed a novel framework deep learning algorithm for the analysis of ECG data that can represent the signal in a convenient form for evaluating different tasks such as ECG signal recognition and heartbeat irregularity identification [99].”