Machine learning evaluation of LV outflow obstruction in hypertrophic cardiomyopathy using three-chamber cardiovascular magnetic resonance

Abstract

Left ventricular outflow tract obstruction (LVOTO) is common in hypertrophic cardiomyopathy (HCM), but relationships between anatomical metrics and obstruction are poorly understood. We aimed to develop machine learning methods to evaluate LVOTO in HCM patients and quantify relationships between anatomical metrics and obstruction. This retrospective analysis of 1905 participants of the HCM Registry quantified 11 anatomical metrics derived from 14 landmarks automatically detected on the three-chamber long axis cine CMR images. Linear and logistic regression was used to quantify strengths of relationships with the presence of LVOTO (defined by resting Doppler pressure drop of > 30 mmHg), using the area under the receiver operating characteristic (AUC). Intraclass correlation coefficients between the network predictions and three independent observers showed similar agreement to that between observers. The distance from anterior mitral valve leaflet tip to basal septum (AML-BS) was most highly correlated with Doppler pressure drop (R2 = 0.19, p < 10–5). Multivariate stepwise regression found the best predictive model included AML-BS, AML length to aortic valve diameter ratio, AML length to LV width ratio, and midventricular septal thickness metrics (AUC 0.84). Excluding AML-BS, metrics grouped according to septal hypertrophy, LV geometry, and AML anatomy each had similar associations with LVOTO (AUC 0.71, 0.71, 0.68 respectively, p = ns), significantly less than their combination (AUC 0.77, p < 0.05 for each). Anatomical metrics derived from a standard three-chamber CMR cine acquisition can be used to highlight risk of LVOTO, and suggest further investigation if necessary. A combination of geometric factors is required to provide the best risk prediction.

Spatial relationship between mitral valve and ventricular septum assessed by resting echocardiography to diagnose left ventricular outflow tract obstruction in hypertrophic cardiomyopathy

“…Interestingly, a recent retrospective study applying machine learning algorithms in a large dataset of HCM patients with available cardiac magnetic resonance imaging reported concordant results. 28 They found that among several anatomical parameters, distance between anterior mitral valve leaflet tip and basal septum best predicted presence of resting LVOTO. Our study further extends these findings involving also patients with provocable LVOTO and providing an easily applicable echocardiographic parameter.…”

Left ventricular anatomy in obstructive hypertrophic cardiomyopathy: beyond basal septal hypertrophy

“…Therefore, we have made available for the clinical and scientific community such metrics in the form of a web service (hosted at the Cardiac Atlas Project, www.cardiacatlas.org ). Although recent work has shown the potential of simpler metrics derived from three-chamber longitudinal views of the heart for the assessment of LVOTO in HCM, 36 our 3D analysis provides more information and can be similarly integrated into the clinical workflow in a fully automated manner. There is an opportunity to explore the assessment of the presented 3D LV phenotypes with data already available in current clinical practice.…”

Imaging of left ventricular outflow tract obstruction in hypertrophic cardiomyopathy

“…Multimodality imaging is essential in confirming the diagnosis of HCM, the presence of LV outflow tract obstruction (LVOTO), and risk stratification for sudden cardiac death [4]. Machine learning evaluation of LV wall thickness and LVOTO is evolving [5,6].…”