Skip to main content

Table 1 Descriptive features of the dataset. For all the numerical variables median and (1st, 3rd quartile) are reported. We compared “Censored” and “Event” populations with Mann–Whitney and Chi-squared tests, respectively for continuous variables and gender. All comparisons were significant (p-value < 0.001)

From: Comparison of discrimination and calibration performance of ECG-based machine learning models for prediction of new-onset atrial fibrillation

 

Censored

Event

Overall

Age (years)

65 (52, 75)

74 (67, 80)

67 (54, 76)

Gender (Male, %)

49

58

50

P axis (degrees)

58 (43, 69)

60 (42, 73)

58 (43, 69)

P onset (msec)

290 (269, 307)

274 (246, 295)

288 (266, 306)

P offset (msec)

407 (388, 422)

391 (361, 413)

406 (385, 422)

PR interval (msec)

163 (148, 182)

176 (157, 199)

164 (149, 184)

QRS axis (degrees)

37 (1, 64)

16 (-22, 53)

35 (-2, 63)

QRS onset (msec)

453 (449, 458)

451 (445, 457)

453 (449, 458)

QRS offset (msec)

550 (543, 558)

551 (545, 563)

550 (543, 558)

QT interval corrected (msec)

408 (395, 424)

420 (404, 439)

409 (396, 426)

T axis (degrees)

54 (36, 68)

59 (34, 78)

55 (36, 69)

T offset (msec)

841 (820, 863)

853 (829, 878)

842 (821, 865)

Heart rate (beats/min)

71 (62, 80)

69 (61, 78)

70 (62, 80)