The adult ECGs were selected from four studies performed in The Netherlands: the Rotterdam Study, the PREVEND Study, the Utrecht Health Project, and the Leiden University dataset. Those with a history of myocardial infarction, heart failure, coronary bypass surgery, coronary angioplasty, or pacemaker implantation were excluded. Other exclusion criteria were hypertension and diabetes mellitus. In such a cardiologically normal population no ECG should be excluded because of its unusual appearance, but, as is usually done, we made an exception for complete right- or left bundle-branch block, Wolf-Parkinson-White pattern, atrial fibrillation or atrial flutter. All ECGs were also visually checked for correct waveform recognition by the ECG computer program. Those with waveform recognition errors (mainly due to excessive noise) were also removed from the dataset. Electrode interchanges were detected by a special feature in MEANS. After visual confirmation these ECGs were also discarded. This left a total of 13,354 participants for analysis. Figure 1 shows the age and sex distribution.

Figure 1. Adult study population



The pediatric population consisted of 1,912 children aged 11 days to 16 years, recruited at three child health centres, three primary schools, and a secondary school in Rotterdam, The Netherlands. Children with previously known cardiovascular abnormalities were excluded. Figure 2 shows the age and sex distribution of the study population.

Figure 2. Pediatric study population