In Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography
BACKGROUND : Echocardiography provides complex data on cardiac function that can be integrated into patterns of dysfunction related to the severity of cardiac disease. The aim is to demonstrate the feasibility of applying machine learning (ML) to automate the integration of echocardiographic data from the whole cardiac cycle, and to automatically recognize patterns in velocity profiles and deformation curves, allowing for identification of functional phenotypes.
METHODS : An echocardiogram was performed in 189 clinically managed hypertensive patients, and 97 non-hypertensive healthy individuals. Speckle-tracking analysis of the left ventricle (LV) and atrium was performed and deformation curves extracted. The aortic and mitral blood pool pulsed-wave (PW) Doppler and the mitral annular tissue PW Doppler velocity profiles were obtained. These whole cardiac cycle deformation and velocity curves were used as the ML input. Unsupervised ML was used to create a representation of hypertensive patients in a virtual space where patients are positioned based on the similarity of their integrated whole cardiac cycle echo data. Regression methods were used to explore patterns of echocardiographic traces within this virtual ML-derived space, while clustering was used to define phenogroups.
RESULTS : The algorithm captured different patterns in tissue/blood-pool velocity and deformation profiles, and integrated the findings, yielding phenotypes related to normal cardiac function and others to advanced remodeling associated with pressure overload in hypertension. The addition of non-hypertensive individuals into the ML-derived space confirmed the interpretation of normal and remodeled phenotypes.
CONCLUSIONS : ML-based pattern recognition is feasible from echocardiographic data obtained during the whole cardiac cycle. Automated algorithms can consistently capture patterns in velocity and deformation data, and, on the basis of these patterns, group patients into interpretable, clinically comprehensive phenogroups that describe structural and functional remodeling. Automated pattern recognition may potentially aid interpretation of imaging data and diagnostic accuracy.
Loncaric Filip, Castellote Pablo-Miki Marti, Sanchez-Martinez Sergio, Fabijanovic Dora, Nunno Loredana, Mimbrero Maria, Sanchis Laura, Doltra Adelina, Montserrat Silvia, Cikes Maja, Crispi Fatima, Piella Gema, Sitges Marta, Bijnens Bart
arterial hypertension, clustering, machine learning, remodeling, speckle tracking