Study flow. Of the 4965 participants of the population-based Characteristics and Course of Heart Failure Stages A–B and Determinants of Progression (STAAB) Cohort Study, 4859 had valid echocardiograms and entered the Automatisierte Vermessung der Echokardiographie project. All echocardiograms had been read at the time point of scanning (human on-site measurement). At project start, the stored pseudonymized echocardiography images of the STAAB participants were imported into the TomtecArena®. Subsequently, all scans were read by the original detector. Further, trained and internally certified personnel of the Academic Core Lab Ultrasound-based Cardiovascular Imaging at the Comprehensive Heart Failure Center, Würzburg, Germany, performed measurements in all echocardiograms, serving as the human referent in further analyses. The STAAB population was then divided into three distinct subgroups using a random algorithm applying a 4:1:1 ratio. The respective group size allowed for a larger training pool (n = 3226), and two equally sized validation pools. The training pool served for machine-learning-based training of the original detector. Images of the first 250 STAAB participants allocated to validation Pool 1 were measured repetitively by observers of different experience, blinded to the results of the other observers (repeat human measurements). The remaining images of validation Pool 1 served for measurements of the re-trained detector. The validation Pool 2 was not used in the current analysis.