Deep Neural Networks Can Predict New-Onset Atrial Fibrillation From the 12-Lead ECG and Help Identify Those at Risk of Atrial Fibrillation-Related Stroke.

Autor:	Raghunath S; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Pfeifer JM; Heart and Vascular Center, Evangelical Hospital, Lewisburg, PA (J.M.P.)., Ulloa-Cerna AE; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Nemani A; Tempus Labs Inc, Chicago, IL (A.N., T.C., K.W.J., N.Z., A.H.)., Carbonati T; Tempus Labs Inc, Chicago, IL (A.N., T.C., K.W.J., N.Z., A.H.)., Jing L; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., vanMaanen DP; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Hartzel DN; Phenomic Analytics and Clinical Data Core (D.N.H., B.F.L., D.B.R., J.B.L.), Geisinger, Danville, PA., Ruhl JA; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Lagerman BF; Phenomic Analytics and Clinical Data Core (D.N.H., B.F.L., D.B.R., J.B.L.), Geisinger, Danville, PA., Rocha DB; Phenomic Analytics and Clinical Data Core (D.N.H., B.F.L., D.B.R., J.B.L.), Geisinger, Danville, PA., Stoudt NJ; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Schneider G; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA., Johnson KW; Tempus Labs Inc, Chicago, IL (A.N., T.C., K.W.J., N.Z., A.H.)., Zimmerman N; Tempus Labs Inc, Chicago, IL (A.N., T.C., K.W.J., N.Z., A.H.)., Leader JB; Phenomic Analytics and Clinical Data Core (D.N.H., B.F.L., D.B.R., J.B.L.), Geisinger, Danville, PA., Kirchner HL; Department of Population Health Sciences (H.L.K.), Geisinger, Danville, PA., Griessenauer CJ; Department of Vascular and Endovascular Neurosurgery (C.J.G.), Geisinger, Danville, PA.; Research Institute of Neurointervention, Paracelsus Medical University, Salzburg, Austria (C.J.G.)., Hafez A; Tempus Labs Inc, Chicago, IL (A.N., T.C., K.W.J., N.Z., A.H.)., Good CW; Heart Institute (C.W.G., B.K.F., C.M.H.), Geisinger, Danville, PA.; Heart and Vascular Institute at University of Pittsburgh Medical Center Hamot, Erie, PA (C.W.G.)., Fornwalt BK; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA.; Heart Institute (C.W.G., B.K.F., C.M.H.), Geisinger, Danville, PA.; Department of Radiology (B.K.F.), Geisinger, Danville, PA., Haggerty CM; Department of Translational Data Science and Informatics (S.R., A.E.U.-C., L.J., D.P.v.M, J.A.R., N.J.S., G.S., B.K.F., C.M.H.), Geisinger, Danville, PA.; Heart Institute (C.W.G., B.K.F., C.M.H.), Geisinger, Danville, PA.
Jazyk:	angličtina
Zdroj:	Circulation [Circulation] 2021 Mar 30; Vol. 143 (13), pp. 1287-1298. Date of Electronic Publication: 2021 Feb 16.
DOI:	10.1161/CIRCULATIONAHA.120.047829
Abstrakt:	Background: Atrial fibrillation (AF) is associated with substantial morbidity, especially when it goes undetected. If new-onset AF could be predicted, targeted screening could be used to find it early. We hypothesized that a deep neural network could predict new-onset AF from the resting 12-lead ECG and that this prediction may help identify those at risk of AF-related stroke. Methods: We used 1.6 M resting 12-lead digital ECG traces from 430 000 patients collected from 1984 to 2019. Deep neural networks were trained to predict new-onset AF (within 1 year) in patients without a history of AF. Performance was evaluated using areas under the receiver operating characteristic curve and precision-recall curve. We performed an incidence-free survival analysis for a period of 30 years following the ECG stratified by model predictions. To simulate real-world deployment, we trained a separate model using all ECGs before 2010 and evaluated model performance on a test set of ECGs from 2010 through 2014 that were linked to our stroke registry. We identified the patients at risk for AF-related stroke among those predicted to be high risk for AF by the model at different prediction thresholds. Results: The area under the receiver operating characteristic curve and area under the precision-recall curve were 0.85 and 0.22, respectively, for predicting new-onset AF within 1 year of an ECG. The hazard ratio for the predicted high- versus low-risk groups over a 30-year span was 7.2 (95% CI, 6.9-7.6). In a simulated deployment scenario, the model predicted new-onset AF at 1 year with a sensitivity of 69% and specificity of 81%. The number needed to screen to find 1 new case of AF was 9. This model predicted patients at high risk for new-onset AF in 62% of all patients who experienced an AF-related stroke within 3 years of the index ECG. Conclusions: Deep learning can predict new-onset AF from the 12-lead ECG in patients with no previous history of AF. This prediction may help identify patients at risk for AF-related strokes.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu