Related Experiment Video
Updated: Aug 26, 2025
Morphological and Functional Assessment of the Right Ventricle Using 3D Echocardiography
Published on: October 28, 2020
Machine learning evaluation of LV outflow obstruction in hypertrophic cardiomyopathy using three-chamber
Manisha Sahota1, Sepas Ryan Saraskani1, Hao Xu1
1Department of Biomedical Engineering, King's College London, 1 Lambeth Palace Rd, London, SE1 7EU, UK.
Insights
Machine learning models can predict left ventricular outflow tract obstruction (LVOTO) in hypertrophic cardiomyopathy (HCM) using cardiac MRI (CMR) anatomical metrics. The distance from the mitral valve to the septum best predicts obstruction, with combined metrics offering superior risk assessment.
Area of Science:
- Cardiology
- Medical Imaging
- Machine Learning
Background:
- Left ventricular outflow tract obstruction (LVOTO) is a frequent complication of hypertrophic cardiomyopathy (HCM).
- The precise relationship between anatomical characteristics and LVOTO severity in HCM remains incompletely understood.
- Accurate assessment of LVOTO is crucial for patient management and risk stratification.
Purpose of the Study:
- To develop and validate machine learning (ML) methods for evaluating LVOTO in HCM patients.
- To quantify the associations between specific anatomical metrics derived from cardiac magnetic resonance (CMR) imaging and the degree of LVOTO.
- To identify key anatomical predictors of LVOTO in HCM.
Main Methods:
- Retrospective analysis of 1905 HCM Registry participants.
- Automatic detection of 14 landmarks on three-chamber cine CMR images to derive 11 anatomical metrics.
- Linear and logistic regression models to assess relationships with LVOTO (defined by Doppler pressure drop > 30 mmHg).
- Evaluation of ML model performance using the area under the receiver operating characteristic curve (AUC).
Main Results:
- The distance from the anterior mitral valve leaflet tip to the basal septum (AML-BS) showed the strongest correlation with Doppler pressure drop (R² = 0.19).
- A multivariate stepwise regression model incorporating AML-BS, AML length to aortic valve diameter ratio, AML length to LV width ratio, and midventricular septal thickness achieved an AUC of 0.84.
- Combinations of anatomical metrics, particularly those related to septal hypertrophy, LV geometry, and AML anatomy, significantly improved LVOTO prediction compared to individual groups.
Conclusions:
- Standard three-chamber CMR cine acquisitions provide valuable anatomical metrics for assessing LVOTO risk in HCM.
- A combination of geometric factors, including AML-BS, offers the best predictive capability for LVOTO.
- These findings support the use of CMR-derived anatomical data to identify HCM patients who may require further investigation for LVOTO.
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.
