Search

Background: Deep learning interpretation of echocardiographic images may facilitate automated assessment of cardiac structure and function., Objectives: We developed a deep learning model to interpret echocardiograms and examined the association of deep learning-derived echocardiographic measures with incident outcomes., Methods: We trained and validated a 3-dimensional convolutional neural network model for echocardiographic view classification and quantification of left atrial dimension, left ventricular wall thickness, chamber diameter, and ejection fraction. The training sample comprised 64,028 echocardiograms (n = 27,135) from a retrospective multi-institutional ambulatory cardiology electronic health record sample. Validation was performed in a separate longitudinal primary care sample and an external health care system data set. Cox models evaluated the association of model-derived left heart measures with incident outcomes., Results: Deep learning discriminated echocardiographic views (area under the receiver operating curve >0.97 for parasternal long axis, apical 4-chamber, and apical 2-chamber views vs human expert annotation) and quantified standard left heart measures (R 2 range = 0.53 to 0.91 vs study report values). Model performance was similar in 2 external validation samples. Model-derived left heart measures predicted incident heart failure, atrial fibrillation, myocardial infarction, and death. A 1-SD lower model-left ventricular ejection fraction was associated with 43% greater risk of heart failure (HR: 1.43; 95% CI: 1.23-1.66) and 17% greater risk of death (HR: 1.17; 95% CI: 1.06-1.30). Similar results were observed for other model-derived left heart measures., Conclusions: Deep learning echocardiographic interpretation accurately quantified standard measures of left heart structure and function, which in turn were associated with future clinical outcomes. Deep learning may enable automated echocardiogram interpretation and disease prediction at scale., Competing Interests: Funding Support and Author Disclosures Dr Lau is supported by the National Institutes of Health (NIH) (K23-HL159243) and the American Heart Association (AHA) (853922). Dr Al-Alusi is supported by the NIH (T32-HL007208). Dr Ellinor is supported by the NIH (1R01HL092577, K24HL105780), AHA (18SFRN34110082), Foundation Leducq (14CVD01), and by MAESTRIA (965286). Dr Lubitz was supported by NIH grants 1R01HL139731 and R01HL157635, and AHA 18SFRN34250007 during this work. Dr Ho is supported by the NIH (R01 HL134893, R01 HL140224, R0 1HL160003, and K24 HL153669). Dr Lau has consulted for Roche Diagnostics. Dr Di Achille is a full-time employee of Google as of June 2022. Ms Kopparapu is a full-time employee of DeepMind as of August 2022. Dr Batra is a full-time employee of Flagship Pioneering as of January 2023. Dr Batra has received sponsored research support from Bayer AG and IBM; and has consulted for Novartis and Prometheus Biosciences. Dr Ellinor has received sponsored research support from Bayer AG and IBM Health; and has served on advisory boards or consulted for Bayer AG, Quest Diagnostics, MyoKardia, and Novartis. Dr Picard has consulted for Vertex. Dr Ho has received past sponsored research support from Bayer AG. Dr Lubitz is a full-time employee of Novartis as of July 2022; has received sponsored research support from Bristol Myers Squibb, Pfizer, Boehringer Ingelheim, Fitbit, Medtronic, Premier, and IBM; and has consulted for Bristol Myers Squibb, Pfizer, Blackstone Life Sciences, and Invitae. All other authors have reported that they had no relationships relevant to the contents of this paper to disclose., (Copyright © 2023 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)