Deep learning on resting electrocardiogram to identify impaired heart rate recovery

Q: What is the clinical evidence from this study?

Study design: Observational. Population: Impaired heart rate recovery. Intervention: Deep learning on resting ECG. Primary outcome: Future clinical outcomes, including new-onset DM and all-cause mortality.

Nathaniel Diamant; Paolo Di Achille; Lu‐Chen Weng; Emily S. Lau; Shaan Khurshid; Sam Friedman; Christopher Reeder; Pulkit Singh; Xin Wang; Gopal Sarma; Mercedeh Ghadessi; Johanna Mielke; Eren Elçi; Ivan Kryukov; Hanna M. Eilken; Andrea Derix; Patrick T. Ellinor; Christopher D. Anderson; Anthony Philippakis; Puneet Batra; Steven A. Lubitz; Jennifer E. Ho

doi:10.1016/j.cvdhj.2022.06.001

Key result

Deep learning-derived heart rate recovery from resting ECG is linked to new-onset DM and mortality.

Figures

Why the study?

Postexercise heart rate recovery reflects cardiac autonomic function and predicts adverse outcomes, prompting investigation into whether deep learning on resting ECG can identify impaired heart rate recovery.

Does a deep learning model applied to resting ECG predict heart rate recovery and future risk of adverse clinical outcomes in adults?

Study Design

Type

Observational

Structured PICO

Does a deep learning model applied to resting ECG predict heart rate recovery and future risk of adverse clinical outcomes in adults?

P

Population

56,793 adults aged 40-69 years from the UK Biobank who underwent exercise testing with ECG monitoring, mean age 57.1, 49% male. Excluded individuals with missing data, nonphysiologic HRR, or ECG artifact.

I

Intervention

Deep learning model (1-dimensional convolutional neural network based on DenseNet) applied to 10-second segments of lead I from pretest resting ECG to predict heart rate recovery (HRRpred).

O

Outcome

Incident diabetes mellitus, congestive heart failure, cardiovascular disease (myocardial infarction, heart failure, or stroke), and all-cause mortality over a median follow-up of 9.9 years.hard clinical

Implication

A deep learning model applied to resting ECG can estimate heart rate recovery, which independently predicts future risk of diabetes, heart failure, and mortality.

Abstract

Background and Objective: Postexercise heart rate recovery (HRR) is an important indicator of cardiac autonomic function and abnormal HRR is associated with adverse outcomes. We hypothesized that deep learning on resting electrocardiogram (ECG) tracings may identify individuals with impaired HRR. Methods: in genome-wide association analysis. Results: included known heart rate, cardiac conduction system, cardiomyopathy, and metabolic trait loci. Conclusion: Deep learning-derived estimates of HRR using resting ECG independently associated with future clinical outcomes, including new-onset DM and all-cause mortality. Inferring postexercise heart rate response from a resting ECG may have potential clinical implications and impact on preventive strategies warrants future study.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Diamant et al. (2022) conducted an observational in Impaired heart rate recovery. Deep learning on resting ECG was evaluated on Future clinical outcomes, including new-onset DM and all-cause mortality. Deep learning-derived estimates of heart rate recovery using resting ECG were independently associated with future clinical outcomes, including new-onset diabetes mellitus and all-cause mortality.

synapsesocial.com/papers/6a0d50f8cae7912d2fa4e633 https://doi.org/10.1016/j.cvdhj.2022.06.001

View Full Paper

Ask AI

Bookmark

Share

Deep learning on resting electrocardiogram to identify impaired heart rate recovery

Figures

Study Design

Structured PICO

Implication

Abstract

Discussion

Cite This Study

Deep learning on resting electrocardiogram to identify impaired heart rate recovery

Figures

Study Design

Structured PICO

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading