High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy With Cardiovascular Deep Learning

Abstract

IMPORTANCEEarly detection and characterization of increased left ventricular (LV) wall thickness can markedly impact patient care but is limited by under-recognition of hypertrophy, measurement error and variability, and difficulty differentiating causes of increased wall thickness, such as hypertrophy, cardiomyopathy, and cardiac amyloidosis.OBJECTIVE To assess the accuracy of a deep learning workflow in quantifying ventricular hypertrophy and predicting the cause of increased LV wall thickness. DESIGN, SETTINGS, AND PARTICIPANTSThis cohort study included physician-curated cohorts from the Stanford Amyloid Center and Cedars-Sinai Medical Center (CSMC) Advanced Heart Disease Clinic for cardiac amyloidosis and the Stanford Center for Inherited Cardiovascular Disease and the CSMC Hypertrophic Cardiomyopathy Clinic for hypertrophic cardiomyopathy from January 1, 2008, to December 31, 2020. The deep learning algorithm was trained and tested on retrospectively obtained independent echocardiogram videos from Stanford Healthcare, CSMC, and the Unity Imaging Collaborative. MAIN OUTCOMES AND MEASURES The main outcome was the accuracy of the deep learning algorithm in measuring left ventricular dimensions and identifying patients with increased LV wall thickness diagnosed with hypertrophic cardiomyopathy and cardiac amyloidosis. RESULTS The study included 23 745 patients: 12 001 from Stanford Health Care (6509 [54.2%] female; mean [SD] age, 61.6 [17.4] years) and 1309 from CSMC (808 [61.7%] female; mean [SD] age, 62.8 [17.2] years) with parasternal long-axis videos and 8084 from Stanford Health Care (4201 [54.0%] female; mean [SD] age, 69.1 [16.8] years) and 2351 from CSMS (6509 [54.2%] female; mean [SD] age, 69.6 [14.7] years) with apical 4-chamber videos. The deep learning algorithm accurately measured intraventricular wall thickness (mean absolute error [MAE], 1.2 mm; 95% CI, 1.1-1.3 mm), LV diameter (MAE, 2.4 mm; 95% CI, 2.2-2.6 mm),and posterior wall thickness (MAE, 1.4 mm; 95% CI, 1.2-1.5 mm) and classified cardiac amyloidosis (area under the curve [AUC], 0.83) and hypertrophic cardiomyopathy (AUC, 0.98) separately from other causes of LV hypertrophy. In external data sets from independent domestic and international health care systems, the deep learning algorithm accurately quantified ventricular parameters (domestic: R 2 , 0.96; international: R 2 , 0.90). For the domestic data set, the MAE was 1.7 mm (95% CI, 1.6-1.8 mm) for intraventricular septum thickness, 3.8 mm (95% CI, 3.5-4.0 mm) for LV internal dimension, and 1.8 mm (95% CI, 1.7-2.0 mm) for LV posterior wall thickness. For the international data set, the MAE was 1.7 mm (95% CI, 1.5-2.0 mm) for intraventricular septum thickness, 2.9 mm (95% CI, 2.4-3.3 mm) for LV internal dimension, and 2.3 mm (95% CI, 1.9-2.7 mm) for LV posterior wall thickness. The deep learning algorithm accurately detected cardiac amyloidosis (AUC, 0.79) and hypertrophic cardiomyopathy (AUC, 0.89) in the domestic external validation site. CONCLUSIONS AND RELEVANCEIn this cohort study, t...

Magnetic-Resonance-Imaging-Based Left Atrial Strain and Left Atrial Strain Rate as Diagnostic Parameters in Cardiac Amyloidosis

“…In spite of recently impressive findings in terms of high-throughput precision phenotyping of left ventricular hypertrophy by cardiovascular deep learning algorithms, the left atrium has become a target of interest in patients with CA. [ 10 ]. Nevertheless, most studies analyzing the role of the left atrium in CA merely set left atrial dimensions as surrogate parameters of disease activity, severity and chronicity without taking into account left atrial function.…”

Automated severe aortic stenosis detection on single-view echocardiography: A multi-center deep learning study

“…From the studies performed between 2016 and 2020 (n=257,829), a stratified weighted sample of 10,000 studies was drawn that overweighted studies with AS (sampling probability weights of 1 for no AS, 5 for non-severe AS, 50 for severe AS). After removing 3,378 studies with no pixel data, de-identifying video frames, and using an automated view classifier to determine the PLAX view, our final derivation set (training and validation) consisted of 6,021 studies with 22,912 videos (1,269,764 frames) (mean age 70.2 ± 15.7 years, n=2950 (49.0%) women), with mild, moderate, and severe AS in 12.4% (n=747), 8.4% (n=503), and 22.3% (n=1,344) of studies, respectively. A held-out, randomly selected, sample of 1,063 studies from the same period was used for (internal) testing, whereas 2,040 randomly selected scans with a total of 6,530 videos performed between January 1 st 2021 and December 15 th 2021 (mean age 65.7 ± 16.4 years, n=997 (48.9%) women) were used for external testing.…”

“…10 Deep learning algorithms have successfully been applied in echocardiograms, where they have shown promise in detecting left ventricular dysfunction, 11 and left ventricular hypertrophy. 12 With the expanded use of point-of-care ultrasonography, 9 developing user-friendly screening algorithms relying on single 2D echocardiographic views would provide an opportunity to improve AS screening. This is however limited by the lack of carefully curated, labelled datasets, as well as efficient ways to utilize the often noisy real-world data for model development.…”

Deep learning enabled analysis of cardiac sphericity

“…Further, manual phenotyping is time-consuming and not feasible for testing novel phenotypes at larger scale. These challenges can be addressed by recent advances in deep learning and computer vision which allow for high throughput automated measurements of cardiac structures [5][6][7][8] , and may identify previously uncharacterized clinically-relevant phenotypic variation 3,9,10 .…”