Deep learning-based NT-proBNP prediction from the ECG for risk assessment in the community
Author:
Neyazi Meraj123, Bremer Jan P.12, Knorr Marius S.12, Gross Stefan45, Brederecke Jan1, Schweingruber Nils6, Csengeri Dora12, Schrage Benedikt12, Bahls Martin45, Friedrich Nele75, Zeller Tanja12, Felix Stephan45, Blankenberg Stefan12, Dörr Marcus45, Vollmer Marcus85, Schnabel Renate B.12
Affiliation:
1. Department of Cardiology , University Heart & Vascular Center Hamburg, University Medical Center Hamburg-Eppendorf , Hamburg , Germany 2. German Centre for Cardiovascular Research (DZHK), Partner Site Hamburg/Kiel/Luebeck , Hamburg , Germany 3. Department of Genetics , Harvard Medical School , Boston, MA , USA 4. Department of Internal Medicine B , University Medicine Greifswald , Greifswald , Germany 5. German Centre for Cardiovascular Research (DZHK), Partner Site Greifswald , Greifswald , Germany 6. Department of Neurology , University Medical Center Hamburg-Eppendorf , Hamburg , Germany 7. Institute for Clinical Chemistry and Laboratory Medicine, University Medicine Greifswald , Greifswald , Germany 8. Institute of Bioinformatics, University Medicine Greifswald , Greifswald , Germany
Abstract
Abstract
Objectives
The biomarker N-terminal pro B-type natriuretic peptide (NT-proBNP) has predictive value for identifying individuals at risk for cardiovascular disease (CVD). However, it is not widely used for screening in the general population, potentially due to financial and operational reasons. This study aims to develop a deep-learning model as an efficient means to reliably identify individuals at risk for CVD by predicting serum levels of NT-proBNP from the ECG.
Methods
A deep convolutional neural network was developed using the population-based cohort study Hamburg City Health Study (HCHS, n=8,253, 50.9 % women). External validation was performed in two independent population-based cohorts (SHIP-START, n=3,002, 52.1 % women, and SHIP-TREND, n=3,819, 51.2 % women). Assessment of model performance was conducted using Pearson correlation (R) and area under the receiver operating characteristics curve (AUROC).
Results
NT-proBNP was predictable from the ECG (R, 0.566 [HCHS], 0.642 [SHIP-START-0], 0.655 [SHIP-TREND-0]). Across cohorts, predicted NT-proBNP (pNT-proBNP) showed good discriminatory ability for prevalent and incident heart failure (HF) (baseline: AUROC 0.795 [HCHS], 0.816 [SHIP-START-0], 0.783 [SHIP-TREND-0]; first follow-up: 0.669 [SHIP-START-1, 5 years], 0.689 [SHIP-TREND-1, 7.3 years]), comparable to the discriminatory value of measured NT-proBNP. pNT-proBNP also demonstrated comparable results for other incident CVD, including atrial fibrillation, stroke, myocardial infarction, and cardiovascular death.
Conclusions
Deep learning ECG algorithms can predict NT-proBNP concentrations with high diagnostic and predictive value for HF and other major CVD and may be used in the community to identify individuals at risk. Long-standing experience with NT-proBNP can increase acceptance of such deep learning models in clinical practice.
Funder
HORIZON EUROPE European Innovation Council
Publisher
Walter de Gruyter GmbH
Subject
Biochemistry (medical),Clinical Biochemistry,General Medicine
Reference36 articles.
1. Zheng, A, Cowan, E, Mach, L, Adam, RD, Guha, K, Cowburn, PJ, et al.. Characteristics and outcomes of patients with suspected heart failure referred in line with National Institute for Health and Care Excellence guidance. Heart 2020;106:1579–85. https://doi.org/10.1136/heartjnl-2019-316511. 2. O’Neal, WT, Sandesara, PB, Claxton, JS, MacLehose, RF, Chen, LY, Bengtson, LGS, et al.. Provider specialty, anticoagulation prescription patterns, and stroke risk in atrial fibrillation. J Am Heart Assoc 2018;7:e007943. https://doi.org/10.1161/jaha.117.007943. 3. Bottle, A, Kim, D, Aylin, P, Cowie, MR, Majeed, A, Hayhoe, B. Routes to diagnosis of heart failure: observational study using linked data in England. Heart 2018;104:600–5. https://doi.org/10.1136/heartjnl-2017-312183. 4. Taylor, CJ, Lay-Flurrie, SL, Ordóñez-Mena, JM, Goyder, CR, Jones, NR, Roalfe, AK, et al.. Natriuretic peptide level at heart failure diagnosis and risk of hospitalisation and death in England 2004–2018. Heart 2022;108:543–9. https://doi.org/10.1136/heartjnl-2021-319196. 5. McDonagh, TA, Metra, M, Adamo, M, Gardner, RS, Baumbach, A, Böhm, M, et al.. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42:3599–726, https://doi.org/10.1093/eurheartj/ehab368.
|
|