Key result
Automated HeartModel yields ~7% lower median LVEF than standard echocardiography in HCM.
Why the study?
Does Philips HeartModel automated assessment accurately measure left ventricular ejection fraction compared to standard echocardiography and cMRI in patients with hypertrophic cardiomyopathy?
Observational (n=50)
No
Does Philips HeartModel automated assessment accurately measure left ventricular ejection fraction compared to standard echocardiography and cMRI in patients with hypertrophic cardiomyopathy?
Automated ML-based LVEF assessment significantly underestimates LVEF and overestimates LV volumes compared to standard echocardiography and cMRI in patients with HCM, suggesting it may not be reliable for guiding cardiac myosin inhibitor therapy.
Answers are grounded in this paper and cite their sources.
May warrant caution using automated LVEF in HCM; hypothesis-generating and requires prospective validation.
Aims Cardiac myosin inhibitors (CMIs) have revolutionized care for patients with obstructive hypertrophic cardiomyopathy (HCM), however they are associated with a risk of systolic dysfunction. Machine learning algorithms might expand access to frequent accurate assessment of left ventricular ejection fraction (LVEF). We assess the performance of a commercial ML-based LVEF model for patients with HCM. Methods and results Single centre prospective study of measurements of left ventricular function by Philips HeartModel® (automated) assessment, echocardiographer assessment (standard), and cardiac magnetic resonance imaging for patients with HCM. Assessments of LVEF, end diastolic volume and end systolic volume were studied across methods. 50 patients with HCM were included. Median age 64 years; 64% male; and 62% had cMRI data for analysis. Median automated LVEF was lower than standard [55.5% (IQR 9) vs. 62.5% (IQR 10), P 0.002, median difference—8% (IQR 14)] and cMRI assessment [55.5% (IQR 9) vs. 68% (IQR 9.5), P < 0.001, median difference—12% (IQR 16)]. Automated assessment traced larger EDVs and ESVs compared with standard 2D tracings [141 mL (IQR 66) vs. 114 mL (IQR 55), P 0.001, and 64 mL (IQR 35) vs. 41 mL (IQR 25), P < 0.001]. Automated assessment identified 11 (22%) patients as having LVEF < 50% vs. 6 (12%) patients identified by expert imaging assessment. Conclusion For patients with HCM, automated assessments of LV size and function differ significantly from standard assessments, raising concerns about the use of this ML-enabled LVEF software for this patient population and potential application to guiding CMI treatments.
No takes yet. Share an insight, caveat, or question.
Morrissey et al. (2026) conducted an observational in Hypertrophic cardiomyopathy (n=50). Philips HeartModel automated assessment vs. Standard echocardiographer assessment and cMRI was evaluated on Left ventricular ejection fraction (LVEF) (median difference -8%, p=0.002). Automated LVEF assessment using Philips HeartModel yielded lower values than standard echocardiography in patients with hypertrophic cardiomyopathy.
Synapse has enriched 10 related papers. Consider them for comparative context: