ECG-only explainable deep learning algorithm predicts the risk for malignant ventricular arrhythmia in phospholamban cardiomyopathy.

Autor:	van de Leur RR; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands. Electronic address: r.r.vandeleur@umcutrecht.nl., de Brouwer R; Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands., Bleijendaal H; Department of Cardiology, Amsterdam UMC location University of Amsterdam, Amsterdam, The Netherlands; European Reference Network for Rare, Low-Prevalence, or Complex Diseases of the Heart (ERN GUARD-Heart); Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Amsterdam University Medical Center, University of Amsterdam, Amsterdam, The Netherlands., Verstraelen TE; Department of Cardiology, Amsterdam UMC location University of Amsterdam, Amsterdam, The Netherlands; European Reference Network for Rare, Low-Prevalence, or Complex Diseases of the Heart (ERN GUARD-Heart)., Mahmoud B; Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands., Perez-Matos A; Department of Cardiology, St Antonius Hospital, Sneek, The Netherlands., Dickhoff C; Department of Cardiology, Dijklander Hospital, Hoorn, The Netherlands., Schoonderwoerd BA; Department of Cardiology, Medical Centre Leeuwarden, Leeuwarden, The Netherlands., Germans T; Department of Cardiology, Noordwest Hospital Group, Alkmaar, The Netherlands., Houweling A; Department of Human Genetics, Amsterdam University Medical Center, Amsterdam, The Netherlands., van der Zwaag PA; Department of Genetics, University Medical Center Groningen, Groningen, The Netherlands., Cox MGPJ; Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands., Peter van Tintelen J; European Reference Network for Rare, Low-Prevalence, or Complex Diseases of the Heart (ERN GUARD-Heart); Department of Genetics, University Medical Center Utrecht, Utrecht, The Netherlands., Te Riele ASJM; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands., van den Berg MP; Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands., Wilde AAM; Department of Cardiology, Amsterdam UMC location University of Amsterdam, Amsterdam, The Netherlands; European Reference Network for Rare, Low-Prevalence, or Complex Diseases of the Heart (ERN GUARD-Heart)., Doevendans PA; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands; European Reference Network for Rare, Low-Prevalence, or Complex Diseases of the Heart (ERN GUARD-Heart); Netherlands Heart Institute, Utrecht, The Netherlands; Central Military Hospital, Utrecht, The Netherlands., de Boer RA; Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands; Department of Cardiology, Erasmus Medical Center, Rotterdam, The Netherlands., van Es R; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands.
Jazyk:	angličtina
Zdroj:	Heart rhythm [Heart Rhythm] 2024 Jul; Vol. 21 (7), pp. 1102-1112. Date of Electronic Publication: 2024 Feb 23.
DOI:	10.1016/j.hrthm.2024.02.038
Abstrakt:	Background: Phospholamban (PLN) p.(Arg14del) variant carriers are at risk for development of malignant ventricular arrhythmia (MVA). Accurate risk stratification allows timely implantation of intracardiac defibrillators and is currently performed with a multimodality prediction model. Objective: This study aimed to investigate whether an explainable deep learning-based approach allows risk prediction with only electrocardiogram (ECG) data. Methods: A total of 679 PLN p.(Arg14del) carriers without MVA at baseline were identified. A deep learning-based variational auto-encoder, trained on 1.1 million ECGs, was used to convert the 12-lead baseline ECG into its FactorECG, a compressed version of the ECG that summarizes it into 32 explainable factors. Prediction models were developed by Cox regression. Results: The deep learning-based ECG-only approach was able to predict MVA with a C statistic of 0.79 (95% CI, 0.76-0.83), comparable to the current prediction model (C statistic, 0.83 [95% CI, 0.79-0.88]; P = .054) and outperforming a model based on conventional ECG parameters (low-voltage ECG and negative T waves; C statistic, 0.65 [95% CI, 0.58-0.73]; P < .001). Clinical simulations showed that a 2-step approach, with ECG-only screening followed by a full workup, resulted in 60% less additional diagnostics while outperforming the multimodal prediction model in all patients. A visualization tool was created to provide interactive visualizations (https://pln.ecgx.ai). Conclusion: Our deep learning-based algorithm based on ECG data only accurately predicts the occurrence of MVA in PLN p.(Arg14del) carriers, enabling more efficient stratification of patients who need additional diagnostic testing and follow-up. Competing Interests: Disclosures The UMC Groningen, which employs several of the authors, received research grants and/or fees from AstraZeneca, Abbott, Boehringer Ingelheim, Cardior Pharmaceuticals GmbH, Ionis Pharmaceuticals, Inc, Novo Nordisk, and Roche (outside the submitted work). Rudolf A. de Boer has had speaker engagements with Abbott, AstraZeneca, Bayer, Bristol Myers Squibb, Novartis, and Roche (outside the submitted work). Rutger R. van de Leur and René van Es are cofounders, shareholders, and board members of Cordys Analytics B.V., a spin-off of the UMC Utrecht that has licensed AI-ECG algorithms, not including the algorithm studied in the current manuscript. The UMC Utrecht receives royalties from Cordys Analytics for potential future revenues. Pieter A. Doevendans is founder and shareholder of HeartEye B.V., an ECG-device company. The other authors declare that there is no conflict of interest. (Copyright © 2024 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu