Skip to main content
Journal cover image

A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using single lead electrocardiograms of variable length.

Publication ,  Journal Article
Kamaleswaran, R; Mahajan, R; Akbilgic, O
Published in: Physiol Meas
March 27, 2018

OBJECTIVE: Atrial fibrillation (AF) is a major cause of hospitalization and death in the United States. Moreover, as the average age of individuals increases around the world, early detection and diagnosis of AF become even more pressing. In this paper, we introduce a novel deep learning architecture for the detection of normal sinus rhythm, AF, other abnormal rhythms, and noise. APPROACH: We have demonstrated through a systematic approach many hyperparameters, input sets, and optimization methods that yielded influence in both training time and performance accuracy. We have focused on these properties to identify an optimal 13-layer convolutional neural network (CNN) model which was trained on 8528 short single-lead ECG recordings and evaluated on a test dataset of 3658 recordings. MAIN RESULTS: The proposed CNN architecture achieved a state-of-the-art performance in identifying normal, AF and other rhythms with an average F 1-score of 0.83. SIGNIFICANCE: We have presented a robust deep learning-based architecture that can identify abnormal cardiac rhythms using short single-lead ECG recordings. The proposed architecture is computationally fast and can also be used in real-time cardiac arrhythmia detection applications.

Duke Scholars

Altmetric Attention Stats
Dimensions Citation Stats

Published In

Physiol Meas

DOI

EISSN

1361-6579

Publication Date

March 27, 2018

Volume

39

Issue

3

Start / End Page

035006

Location

England

Related Subject Headings

  • Signal Processing, Computer-Assisted
  • Neural Networks, Computer
  • Humans
  • Heart Rate
  • Electrocardiography
  • Biomedical Engineering
  • Arrhythmias, Cardiac
  • 4003 Biomedical engineering
  • 3208 Medical physiology
  • 1116 Medical Physiology
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Kamaleswaran, R., Mahajan, R., & Akbilgic, O. (2018). A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using single lead electrocardiograms of variable length. Physiol Meas, 39(3), 035006. https://doi.org/10.1088/1361-6579/aaaa9d
Kamaleswaran, Rishikesan, Ruhi Mahajan, and Oguz Akbilgic. “A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using single lead electrocardiograms of variable length.Physiol Meas 39, no. 3 (March 27, 2018): 035006. https://doi.org/10.1088/1361-6579/aaaa9d.
Kamaleswaran, Rishikesan, et al. “A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using single lead electrocardiograms of variable length.Physiol Meas, vol. 39, no. 3, Mar. 2018, p. 035006. Pubmed, doi:10.1088/1361-6579/aaaa9d.
Journal cover image

Published In

Physiol Meas

DOI

EISSN

1361-6579

Publication Date

March 27, 2018

Volume

39

Issue

3

Start / End Page

035006

Location

England

Related Subject Headings

  • Signal Processing, Computer-Assisted
  • Neural Networks, Computer
  • Humans
  • Heart Rate
  • Electrocardiography
  • Biomedical Engineering
  • Arrhythmias, Cardiac
  • 4003 Biomedical engineering
  • 3208 Medical physiology
  • 1116 Medical Physiology