Deep Learning Model of Diastolic Dysfunction Risk Stratifies the Progression of Early-Stage Aortic Stenosis

Abstract

BackgroundThe development and progression of aortic stenosis (AS) from aortic valve (AV) sclerosis is highly variable and difficult to predict.ObjectivesWe investigated whether a previously validated echocardiography-based deep learning (DL) model assessing diastolic dysfunction (DD) could identify the latent risk associated with the development and progression of AS.MethodsWe evaluated 898 participants with AV sclerosis from the Atherosclerosis Risk in Communities (ARIC) cohort study and associated the DL-predicted probability of DD with two endpoints: (1) the new diagnosis of AS and (2) the composite of subsequent mortality or AV interventions. We performed validation in two additional cohorts: 1) patients with mild-to-moderate AS undergoing cardiac magnetic resonance (CMR) imaging and serial echocardiographic assessments (n=50), and (2) patients with AV sclerosis undergoing18F-sodium fluoride (18F-NaF) and18F-fluorodeoxyglucose positron emission tomography (PET) combined with computed tomography (CT) to assess valvular inflammation and calcification (n=18).ResultsIn the ARIC cohort, a higher DL-predicted probability of DD was associated with the development of AS (adjusted HR: 3.482 [2.061 – 5.884], p<0.001) and subsequent mortality or AV interventions (adjusted HR: 7.033 [3.036 – 16.290], p<0.001). The multivariable Cox model (incorporating the DL-predicted probability of DD) derived from the ARIC cohort efficiently predicted the progression of AS (C-index: 0.798 [0.648 – 0.948]) in the CMR cohort. Moreover, the predictions of this multivariable Cox model correlated positively with valvular18F-NaF mean standardized uptake values in the PET/CT cohort (r=0.62, p=0.008).ConclusionsAssessment of DD using DL can stratify the latent risk associated with the progression of early-stage AS.CONDENSED ABSTRACTWe investigated whether DD assessed using DL can predict the progression of early-stage AS. In 898 patients with AV sclerosis, the DL-predicted probability of DD was associated with the development of AS. The multivariable Cox model derived from these patients also predicted the progression of AS in an external cohort of patients with mild-to-moderate AS (n=50). Moreover, the predictions of this model correlated positively with PET/CT-derived valvular18F-NaF uptake in an additional cohort of patients with AV sclerosis (n=18). These findings suggest that assessing DD using DL can stratify the latent risk associated with the progression of early-stage AS.