Utilizing echocardiography and unsupervised machine learning for heart failure risk identification

• Published in: International journal of cardiology Vol. 418; p. 132636
• Main Authors: Simonsen, Jakob Øystein, Modin, Daniel, Skaarup, Kristoffer, Djernæs, Kasper, Lassen, Mats Christian Højbjerg, Johansen, Niklas Dyrby, Marott, Jacob Louis, Jensen, Magnus Thorsten, Jensen, Gorm B., Schnohr, Peter, Martínez, Sergio Sánchez, Claggett, Brian Lee, Møgelvang, Rasmus, Biering-Sørensen, Tor
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.01.2025
• Subjects: Adult; Aged; Artificial intelligence; Cluster analysis; Echocardiography; Echocardiography - methods; Female; Follow-Up Studies; Heart failure; Heart Failure - diagnosis; Heart Failure - diagnostic imaging; Heart Failure - epidemiology; Heart Failure - physiopathology; Humans; Longitudinal strain; Male; Middle Aged; Risk Assessment - methods; Unsupervised Machine Learning; Cluster analysis; Heart failure; Unsupervised machine learning; Longitudinal strain; Echocardiography; uML; HCT; LVEF; Artificial intelligence; HF; ML; GLS
• ISSN: 0167-5273; 1874-1754; 1874-1754
• DOI: 10.1016/j.ijcard.2024.132636

Global longitudinal strain (GLS) is recognized as a powerful predictor of heart failure (HF). However, the entire strain curve may entail important prognostic information regarding HF risk that might be undiscovered by only focusing on the peak strain value. The hypothesis of the present study was, that analysis of the entire strain curve using unsupervised machine learning (uML) would reveal novel ventricular deformation patterns capable of predicting incident HF independently of GLS. Longitudinal strain curves from 3710 subjects from the general population without prevalent HF were analyzed using uML. Mean age was 56 years and 43 % were male. During a median follow-up of 5.3 years, 92 subjects (2.5 %) developed HF. The uML algorithm generated a hierarchical clustering tree (HCT) resulting in 10 different clusters. Generally, the strain curves displayed reduced early diastolic strain to peak-strain ratio with an increasing incidence rate of HF. In multivariable Cox regressions, cluster 9 was significantly associated with increased risk of HF when compared to cluster 2–5, and 7–8 [For cluster 3: HR 8.95, 95 %CI: 2.08;38.48, P = 0.003] even though the subjects of cluster 9 were younger, displayed healthier clinical baseline characteristics, and only had slightly reduced GLS. The mean strain curve of cluster 9 displayed an early systolic lengthening followed by a late and reduced contraction specifically related to the basal lateral segment. The unsupervised machine learning algorithm identified unknown strain patterns beyond GLS presumably related to increased risk of HF. Strain curves from echocardiographic examinations of 3710 people were clustered into 10 groups using unsupervised machine learning. Clusters with reduced early diastolic strain to peak-strain ratio had an increased incidence rate of heart failure (Left figure). One cluster (Cluster 9) had a significantly increased risk of heart failure and displayed an early systolic lengthening in the basal lateral segment (Right figure). [Display omitted] •ML clustered subjects by HF risk based on similarities in strain curve patterns.•One cluster showed high HF risk despite young age and near normal GLS.•Findings suggest regional strain curves hold more info than GLS.