Skip to main content

Table 1 Characteristics of subjects selected for the training and test data from the ALLSTAR database

From: Optimal length of R–R interval segment window for Lorenz plot detection of paroxysmal atrial fibrillation by machine learning

 

Training data

Test data

Non-AF

CAF

Non-AF

PAF

n

58

52

52

53

Male, %

36.2

61.5

42.3

62.3

Age, years

61.4 ± 12.5

78.0 ± 8.4

65.7 ± 11.8

68.2 ± 9.5

PAF frequency/24 h

4 (1–14)

Individual PAF length, beat

7 (4–23)

Total PAF duration, min/24 h

0

139.9 (10.3–579.3)

Total PAF beat/24 h

0

12,666 (1263–51,983)

APC/24 h

31 (10–61)

54 (11–150)

316 (88–1849)

VPC/24 h

5 (1–39)

48 (9–196)

10 (1–24)

6 (1–28)

24-h mean RRI, ms

844 ± 105

789 ± 141

873 ± 135

801 ± 119

24-h SD of RRI, ms

138 ± 36

217 ± 57

134 ± 43

185 ± 57

  1. Data are frequency (%), median (IQR), or mean ± SD
  2. AF atrial fibrillation, CAF chronic AF, PAF paroxysmal AF, RRI R–R interval, SR sinus rhythm, APC atrial premature contraction, VPC ventricular premature contraction