Table 1 Demographics and sleep statistics of participants in the two data sets used in the study.

From: A deep transfer learning approach for wearable sleep stage classification with photoplethysmography

Parameter

Siesta data

Eindhoven data

N

292 participants (126 females, 43.2%), 584 recordings (252 females, 43.2%)

60 participants (26 females, 43.3%), 101 recordings (48 females, 47.5%)

Age (year)

51.5 (17.3), 20.0−95.0

51.1 (7.9), 41.0−66.0

BMI (kg/m2)

25.6 (4.5), 16.5−43.3

25.6 (3.9), 17.5−36.2

TIB (hour)

8.0 (0.5), 5.8−9.6

7.9 (0.7), 6.4−10.3

SE (%)

80.8 (12.8), 14.6−99.1

85.0 (9.8), 36.0−96.6

N1 sleep (%)

13.1 (8.4), 2.4−77.1

10.7 (5.0), 3.0−30.6

N2 sleep (%)

53.8 (8.8), 13.6−78.8

41.7 (8.7), 22.2−66.6

N3 sleep (%)

13.8 (8.4), 0.0−44.5

26.2 (8.7), 10.3−47.3

REM sleep (%)

18.2 (5.9), 0.0−34.8

21.4 (5.9), 9.2−38.2

  1. Note: Sleep statistics are computed based on the sleep stage annotation of the data set. Except for N, results are presented as mean (standard deviation), range. The percentages of N1, N2, N3, and REM sleep were normalized to total sleep time (excluding wake time). BMI body mass index, TIB time in bed, SE sleep efficiency.
Back to article page