European Heart Journal
Heart-recording model detects and predicts a heart muscle disease
Its scores for distinguishing people with and without the condition ranged from 0.87 to 0.97, outperforming traditional heart-recording measures.Its scores for distinguishing people with and without the condition ranged from 0.87 to 0.97, outperforming traditional heart-recording measures.
Cohort · 20,358 peopleBiomarkers
Who was studied20,358 patients undergoing ECG and echocardiography
What changedThe AI model detected heart muscle abnormalities more accurately than traditional ECG measures (AUROC .90 versus .75)
The catchTraining labels captured heart enlargement and dysfunction without systematically confirming their cause
