Key result
A machine learning model using clinical and ECG features detected abnormal left ventricular geometry with 87% accuracy, 97% sensitivity, and an AUC of 0.91.
Why the study?
Cardiac remodeling is an important aspect of cardiovascular disease progression, and machine learning techniques were investigated to detect abnormal left ventricular geometry before left ventricular hypertrophy onset in patients without established cardiovascular disease.
Can machine learning applied to clinical and ECG features accurately detect abnormal left ventricular geometry in patients without established cardiovascular disease?
Observational (n=528)
Can machine learning applied to clinical and ECG features accurately detect abnormal left ventricular geometry in patients without established cardiovascular disease?
Effect estimate: AUC 0.91
Machine learning applied to standard ECG and clinical parameters can accurately detect early abnormal left ventricular geometry before the onset of overt hypertrophy.
No takes yet. Share an insight, caveat, or question.
May support ECG-based ML screening for abnormal LV geometry; leaves open clinical utility pending prospective validation.
Angelaki et al. (2021) conducted an observational in No established cardiovascular disease (n=528). Machine learning model using clinical and ECG features was evaluated on Detection of abnormal left ventricular geometry (distinguishing normal geometry from concentric remodeling and left ventricular hypertrophy combined) (AUC 0.91). A machine learning model using clinical and ECG features detected abnormal left ventricular geometry with 87% accuracy, 97% sensitivity, and an AUC of 0.91.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: