Table 4.
Sleep parameters
| Healthy volunteers test set* | ||||||||
|---|---|---|---|---|---|---|---|---|
| Nightstand radar (n = 23) | Ceiling radar (n = 21) | Actigraphy (n = 23) | ||||||
| Model type† | Variable‡ | PSG§ (n = 23) | Mean (SD) | Cohen's D | Mean (SD) | Cohen's D | Mean (SD) | Cohen's D |
| Four past, two future | TST [min] | 434.4 (18.2) | 437.4 (17.2) | 0.17 | 438.3 (22.6) | 0.24 | 433.2 (20.1) | −0.06 |
| SOL [min] | 10.5 (6.2) | 12.6 (6.4)¶ | 0.33 | 11.0 (6.0) | −0.01 | 13.7 (6.6)¶ | 0.5 | |
| WASO [min] | 17.3 (11.2) | 15.1 (13.1) | −0.18 | 17.5 (17.8) | −0.05 | 19.3 (17.3) | 0.13 | |
| SE [%] | 94.0 (2.6) | 94.1 (2.9) | 0.01 | 93.9 (4.3) | 0.06 | 93.0 (4.0) | −0.31 | |
| NW [num] | 17.6 (4.7) | 9.8 (3.9)# | −1.8 | 8.0 (4.7) # | −2.12 | 12.8 (5.5)¶ | −0.93 | |
| Five past, zero future (real time) | TST [min] | 434.4 (18.2) | 436.9 (16.4) | 0.15 | 437.3 (21.7) | 0.2 | 433.1 (19.3) | −0.07 |
| SOL [min] | 10.5 (6.2) | 13.0 (6.0)¶ | 0.41 | 12.4 (6.4) | 0.21 | 14.5 (6.5)¶ | 0.63 | |
| WASO [min] | 17.3 (11.2) | 15.8 (12.1) | −0.13 | 17.7 (17.0) | −0.04 | 19.2 (15.8) | 0.14 | |
| SE [%] | 94.0 (2.6) | 93.9 (2.6) | −0.07 | 93.6 (4.2) | −0.03 | 92.8 (3.8) | −0.37 | |
| NW [num] | 17.6 (4.7) | 12.3 (4.9)# | −1.09 | 10.0 (5.3)# | −1.57 | 15.2 (5.7) | −0.46 | |
| Patients with sleep disorders|| | ||||||||
| PSG (n = 28) | Nightstand radar (n = 28) | Actigraph (n = 28) | ||||||
| Model type | Variable | Mean (SD) | Cohen's D | Mean (SD) | Cohen's D | |||
| Four past, two future | TST [min] | 386.7 (73.3) | 386.5 (85.2) | 0 | 372.0 (59.3) | −0.22 | ||
| SOL [min] | 13.0 (14.9) | 10.0 (12.4) | −0.22 | 14.2 (16.1) | 0.07 | |||
| WASO [min] | 61.3 (55.7) | 64.9 (77.8) | 0.05 | 74.0 (42.6) | 0.26 | |||
| SE [%] | 84.2 (12.2) | 84.7 (16.1) | 0.03 | 81.6 (9.5) | −0.24 | |||
| NW [num] | 24.6 (13.2) | 19.3 (14.0) | −0.39 | 21.6 (10.5) | −0.25 | |||
| Five past, zero future (realtime) | TST [min] | 386.7 (73.3) | 387.2 (85.3) | 0.01 | 374.4 (58.2) | −0.19 | ||
| SOL [min] | 13.0 (14.9) | 10.3 (12.2) | −0.2 | 15.4 (18.1) | 0.14 | |||
| WASO [min] | 61.3 (55.7) | 61.9 (80.8) | 0.01 | 71.2 (39.2) | 0.2 | |||
| SE [%] | 84.2 (12.2) | 85.5 (16.2) | 0.09 | 82.0 (9.0) | −0.2 | |||
| NW [num] | 24.6 (13.2) | 23.1 (15.5) | −0.1 | 23.9 (11.5) | −0.05 | |||
Sleep parameters extracted from manually scored hypnograms (PSG) and from the sleep/wake state sequences resulting from the application of two selected classification models to radar and actigraphy data. Sleep onset latency was calculated from a user-marker input. The significant difference found from paired-sample Student's t-tests of each model/sensor compared to their corresponding PSG recordings are indicated with superscripts, and the standardized effect sizes are reported by Cohen's D.
*n = 12, mean age ± SD: 23.0 ± 3.1 years, 5 male, 4 nights of PSG + actigraphy + two radars per participant. The participants were randomly assigned into a training set for model development (n = 24/22 for nightstand/ceiling), and a testing set for validation (n = 23/21 for nightstand/ceiling).
†The models are defined by the number of preceding (past) and succeeding (future) epochs used to score a present epoch.
‡ SOL, Sleep Onset Latency; TST, Total Sleep Time; WASO, Wake After Sleep Onset; SE, Sleep Efficiency; NW, Number of awakenings.
§PSG, Polysomnography (sleep parameters scored manually, independent of models).
||Ambulatory sleep disorder patients. n = 28, mean age ± SD: 46.25 ± 13.98 years, 19 male.
¶ p < 0.05 Student's t-test, compared to PSG.
# p < 0.001.