Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jul 31;2(1):e000193.
doi: 10.1136/bmjmed-2022-000193. eCollection 2023.

-- The emerging role of artificial intelligence enabled electrocardiograms in healthcare

Arunashis Sau  1   2 , Fu Siong Ng  1   2   3
Affiliations

-- The emerging role of artificial intelligence enabled electrocardiograms in healthcare

Arunashis Sau et al. BMJ Med. .
No abstract available

Keywords: Cardiology.

PubMed Disclaimer

Conflict of interest statement

Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following interests: none.

Figures

Figure 1
Figure 1
Examples of how artificial intelligence, machine learning, and deep learning can be applied to the ECG. Random forest is an ensemble of several uncorrelated decision trees, each using a random subset of features. Each tree makes a classification prediction, the class with the most trees with that prediction is the model’s final prediction. Principle component analysis (PCA) is a form of linear dimensionality reduction that can be applied, for example, to ECG measurements. PCA can be combined with clustering algorithms, such as K-means, to find clusters of patients who may have difference clinical features and outcomes. Neural networks are multilayered non-linear models that can be used to make connections within the input data and determine complex associations between input and label to make classification decisions. A variational autoencoder is a form of unsupervised deep learning where the network learns to compress the input into a latent space distribution before reconstructing and aiming to return to the same signal as the input. Variational autoencoders are often used for non-linear dimensionality reduction or removal of noise from signals. A single layer autoencoder with linear activation functions is almost equivalent to PCA. SCD=sudden cardiac death; SR=sinus rhythm; LBBB=left bundle block; PVC=premature ventricular contraction
Figure 2
Figure 2
Flow of data in a machine learning study. An internal dataset is typically split into three sets: training (Train), validation (Val), and testing (Test). Training and validation sets are used for model development before evaluation on the unseen internal test set. An external dataset is ideally used in addition to the internal test dataset to ensure that the model generalises well to a different population
Figure 3
Figure 3
A depiction of how AI enabled ECG may be used in the near future to guide tailored further investigation and management in both outpatient and emergency settings. Examples of potential applications of AI enabled ECG prediction in each setting are given, with how these predications may guide clinical management
See this image and copyright information in PMC

References

    1. Quer G, Arnaout R, Henne M, et al. . Machine learning and the future of cardiovascular care. Journal of the American College of Cardiology 2021;77:300–13. 10.1016/j.jacc.2020.11.030 - DOI - PMC - PubMed
    1. Goodfellow I, Yoshua B, Courville A. Deep Learning. MIT Press, 2016.
    1. Smith SW, Walsh B, Grauer K, et al. . A deep neural network learning algorithm Outperforms a conventional algorithm for emergency Department electrocardiogram interpretation. J Electrocardiol 2019;52:88–95. 10.1016/j.jelectrocard.2018.11.013 - DOI - PubMed
    1. Ribeiro AH, Ribeiro MH, Paixão GMM, et al. . Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat Commun 2020;11. 10.1038/s41467-020-15432-4 - DOI - PMC - PubMed
    1. Attia ZI, Kapa S, Lopez-Jimenez F, et al. . Screening for cardiac contractile dysfunction using an artificial intelligence-enabled electrocardiogram. Nat Med 2019;25:70–4. 10.1038/s41591-018-0240-2 - DOI - PubMed

LinkOut - more resources