ECG Arrhythmia Classification Using Transfer Learning from 2-Dimensional Deep CNN Features

TL;DR

This paper demonstrates that transfer learning from deep CNNs trained on image data can effectively classify ECG arrhythmias with high accuracy, reducing the need for large ECG-specific datasets.

Contribution

The study introduces a transfer learning approach using deep CNN features from image classification to ECG arrhythmia detection, achieving high accuracy with limited ECG data.

Findings

Achieved 97.23% classification accuracy on nearly 7000 ECG instances.

Used features from deep CNNs trained on image data for ECG classification.

Validated effectiveness of transfer learning in medical signal analysis.

Abstract

Due to the recent advances in the area of deep learning, it has been demonstrated that a deep neural network, trained on a huge amount of data, can recognize cardiac arrhythmias better than cardiologists. Moreover, traditionally feature extraction was considered an integral part of ECG pattern recognition; however, recent findings have shown that deep neural networks can carry out the task of feature extraction directly from the data itself. In order to use deep neural networks for their accuracy and feature extraction, high volume of training data is required, which in the case of independent studies is not pragmatic. To arise to this challenge, in this work, the identification and classification of four ECG patterns are studied from a transfer learning perspective, transferring knowledge learned from the image classification domain to the ECG signal classification domain. It is demonstrated that feature maps learned in a deep neural network trained on great amounts of generic input images can be used as general descriptors for the ECG signal spectrograms and result in features that enable classification of arrhythmias. Overall, an accuracy of 97.23 percent is achieved in classifying near 7000 instances by ten-fold cross validation.