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 Nov 8;5(1):89-96.
doi: 10.1093/ehjdh/ztad070. eCollection 2024 Jan.

Automatic triage of twelve-lead electrocardiograms using deep convolutional neural networks: a first implementation study

Rutger R van de Leur  1 , Meike T G M van Sleuwen  1 , Peter-Paul M Zwetsloot  1 , Pim van der Harst  1 , Pieter A Doevendans  1   2   3 , Rutger J Hassink  1 , René van Es  1
Affiliations

Automatic triage of twelve-lead electrocardiograms using deep convolutional neural networks: a first implementation study

Rutger R van de Leur et al. Eur Heart J Digit Health. .

Abstract

Aims: Expert knowledge to correctly interpret electrocardiograms (ECGs) is not always readily available. An artificial intelligence (AI)-based triage algorithm (DELTAnet), able to support physicians in ECG prioritization, could help reduce current logistic burden of overreading ECGs and improve time to treatment for acute and life-threatening disorders. However, the effect of clinical implementation of such AI algorithms is rarely investigated.

Methods and results: Adult patients at non-cardiology departments who underwent ECG testing as a part of routine clinical care were included in this prospective cohort study. DELTAnet was used to classify 12-lead ECGs into one of the following triage classes: normal, abnormal not acute, subacute, and acute. Performance was compared with triage classes based on the final clinical diagnosis. Moreover, the associations between predicted classes and clinical outcomes were investigated. A total of 1061 patients and ECGs were included. Performance was good with a mean concordance statistic of 0.96 (95% confidence interval 0.95-0.97) when comparing DELTAnet with the clinical triage classes. Moreover, zero ECGs that required a change in policy or referral to the cardiologist were missed and there was a limited number of cases predicted as acute that did not require follow-up (2.6%).

Conclusion: This study is the first to prospectively investigate the impact of clinical implementation of an ECG-based AI triage algorithm. It shows that DELTAnet is efficacious and safe to be used in clinical practice for triage of 12-lead ECGs in non-cardiology hospital departments.

Keywords: Deep learning; Electrocardiography; Implementation; Triage.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: R.R.v.d.L. and R.v.E. are cofounders, shareholders, and board members of Cordys Analytics B.V., a spin-off of the UMC Utrecht that has licenced AI-ECG algorithms, including the algorithm studied in the current manuscript. The UMC Utrecht receives royalties from Cordys Analytics for potential future revenues. P.A.D. is a founder and shareholder of HeartEye B.V., an ECG device company.

Figures

Graphical Abstract
Graphical Abstract
Overview of the study and its outcomes. AUROC, area under the receiver operating curve; CI, confidence interval; ECG, electrocardiogram.
Figure 1
Figure 1
Labelling into triage classes as based on final clinical diagnosis. For few cases where the cardiologist-annotated diagnosis was not clear (e.g. whether specific or non-specific ST abnormalities), the ECG was assessed to determine the appropriate category. When multiple diagnoses were visible on the ECG, the highest triage class was chosen. All other non-acute disorders can be found in Supplementary material online, Figure S1. *ECG changes were defined as ST-segment deviations or T-wave changes associated with ischaemia. ACS, acute coronary syndrome; AV(N)RT, atrioventricular (nodal) re-entry tachycardia; VT, ventricular tachycardia; AV block, atrioventricular block; AF, atrial fibrillation.
Figure 2
Figure 2
Confusion matrix comparing Marquette 12SL and DELTAnet predictions to the clinical triage classes (based on final clinical diagnosis).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, et al. 2017 ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2018;39:119–177. - PubMed
    1. Foo CY, Bonsu KO, Nallamothu BK, Reid CM, Dhippayom T, Reidpath DD, et al. Coronary intervention door-to-balloon time and outcomes in ST-elevation myocardial infarction: a meta-analysis. Heart 2018;104:1362–1369. - PubMed
    1. Terkelsen CJ, Sørensen JT, Maeng M, Jensen LO, Tilsted H-H, Trautner S, et al. System delay and mortality among patients with STEMI treated with primary percutaneous coronary intervention. JAMA 2010;304:763–771. - PubMed
    1. Eslava D, Dhillon S, Berger J, Homel P, Bergmann S. Interpretation of electrocardiograms by first-year residents: the need for change. J Electrocardiol 2009;42:693–697. - PubMed
    1. Cook DA, Oh S-Y, Pusic MV. Accuracy of physicians’ electrocardiogram interpretations. JAMA Intern Med 2020;180:1461. - PMC - PubMed

LinkOut - more resources