| Abstract |
Background: The 2016 American Society of Echocardiography guidelines have been widely used to assess left ventricular diastolic function. However, limitations are present in the current classification system. The aim of this study was to develop a data-driven, unsupervised machine learning approach for diastolic function classification and risk stratification using the left ventricular diastolic function parameters recommended in the 2016 American Society of Echocardiography guidelines; the guideline grading was used as the reference standard.
Methods: Baseline demographics, heart failure hospitalization, and all-cause mortality data were obtained for all adult patients who underwent transthoracic echocardiography at Mayo Clinic Rochester in 2015. Patients with prior mitral valve intervention, congenital heart disease, cardiac transplantation, or cardiac assist device implantation were excluded. Nine left ventricular diastolic function variables (mitral E- and A-wave peak velocities, E/A ratio, deceleration time, medial and lateral annular e' velocities and E/e' ratio, and tricuspid regurgitation peak velocity) were used for an unsupervised machine learning algorithm to identify different phenotype clusters. The cohort average of each variable was used for imputation. Patients were grouped according to the algorithm-determined clusters for Kaplan-Meier survival analysis.
Results: Among 24,414 patients (mean age, 63.6 ± 16.2 years), all-cause mortality occurred in 4,612 patients (18.9%) during a median follow-up period of 3.1 years. The algorithm determined three clusters with echocardiographic measurement characteristics corresponding to normal diastolic function (n = 8,312), impaired relaxation (n = 11,779), and increased filling pressure (n = 4,323), with 3-year cumulative mortality of 11.8%, 19.9%, and 33.4%, respectively (P < .0001). All 10,694 patients (43.8%) classified as indeterminate were reclassified into the three clusters (n = 3,324, n = 5,353, and n = 2,017, respectively), with 3-year mortality of 16.6%, 22.9%, and 34.4%, respectively. The clusters also outperformed guideline-based grade for prognostication (C index = 0.607 vs 0.582, P = .013).
Conclusions: Unsupervised machine learning identified physiologically and prognostically distinct clusters on the basis of nine diastolic function Doppler variables. The clusters can be potentially applied in echocardiography laboratory practice and future clinical trials for simple, replicable diastolic function-related risk stratification.
|
