Skip to main content

Table 1 Baseline characteristics for patients with PAF and non-PAF in a training/validation cohort and a test cohort

From: Machine learning based potentiating impacts of 12-lead ECG for classifying paroxysmal versus non-paroxysmal atrial fibrillation

Feature

Training and validation set

(Hospital A)

Test set

(Hospital B)

PAF

(n = 334)

Non-PAF (n = 437)

p value

PAF

(n = 103)

Non-PAF (n = 497)

p value

Baseline characteristics

      

Age, year

58 (12.1)

58 (10.8)

0.75

69 (10.1)

68 (11.4)

0.42

Female gender, n (%)

91 (27%)

95 (21%)

0.09

51 (49%)

241 (48%)

0.91

Chronic Heart Failure, n (%)

21 (6%)

70 (16%)

 < 0.001

20 (20%)

97 (19%)

0.99

Hypertension, n (%)

164 (49%)

205 (46%)

0.56

59 (57%)

281 (56%)

0.91

Diabetes, n (%)

49 (15%)

79 (18%)

0.2

24 (23%)

121 (24%)

0.89

Cerebrovascular accident (incl. TIA), n (%)

39 (6%)

54 (6%)

0.71

21 (20%)

112 (22%)

0.69

Vascular disease, n (%)

41 (12%)

55 (13%)

0.91

5 (4%)

10 (2%)

0.15

CHA2DS2Vasc Score

1.7 (1.5)

1.73 (1.6)

0.94

2.74 (1.5)

2.75 (1.5)

0.91

Echocardiographic factors

      

Left Atrium size, mm

40.8 (5.8)

44.4 (5.8)

 < 0.001

45.7 (8.1)

53.4 (9.5)

 < 0.001

Left Atrium volume, mm3

35.1 (12.6)

43.3 (13.9)

 < 0.001

42.8 (20.2)

62.6 (37.9)

 < 0.001

Left Ventricular Ejection Fraction, %

65.8 (39.6)

60.8 (8.8)

0.01

56.8 (13.6)

58.8 (24.6)

0.41

  1. Non-paroxysmal atrial fibrillation, non-PAF; paroxysmal atrial fibrillation, PAF