Key result
Machine learning model identifies HFpEF with 0.88 AUC using only demographics and plasma biomarkers.
Why the study?
Does an ensemble machine learning model using matrix-related plasma biomarkers and demographics accurately identify patients with HFpEF?
Observational (n=480)
No
Does an ensemble machine learning model using matrix-related plasma biomarkers and demographics accurately identify patients with HFpEF?
Machine learning algorithms using a limited panel of demographics and matrix-related plasma biomarkers can accurately identify HFpEF without the need for sophisticated echocardiography imaging.
May enable non-invasive HFpEF identification without echocardiography; hypothesis-generating and requires prospective validation before clinical use.
Machine learning algorithms correctly classified patients with heart failure with preserved ejection fraction with over 90% area under receiver-operating-characteristic curves. Classifications using multidomain features (demographics and circulating biomarkers and echo-based ventricle metrics) proved more accurate than previous studies using single-domain features alone. Excitingly, HFpEF diagnoses were generally accurate even without echo-based measurements, demonstrating that such algorithms could provide an early screening tool using blood-based measurements before sophisticated imaging.
No takes yet. Share an insight, caveat, or question.
Ward et al. (2022) conducted an observational in Heart failure with preserved ejection fraction (HFpEF) (n=480). Ensemble machine learning model was evaluated on Area under the receiver-operating-characteristic curve (AUC) for HFpEF diagnosis using demographic and plasma panel data (AUC 0.88). An ensemble machine learning model accurately identified patients with heart failure with preserved ejection fraction, achieving an AUC of 0.88 using only demographic and plasma biomarker data.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: