Temperature variability analysis using wavelets and multiscale entropy in patients with systemic inflammatory response syndrome, sepsis, and septic shock

Table 3 Distribution of wavelet features in the three groups of patients

Wavelet	Group 1	Group 2	Group 3	1 vs 2	1 vs 3	2 vs 3
Features	(SIRS) n= 5	(sepsis) n= 10	(septic shock) n= 7	P	P	P
*sign_m* , raw data with mean value subtracted
WEn(s4)	0.226 (0.19-0.31)	0.159 (0.12-0.21)	0.105 (0.08-0.15)		0.047
WEn(s5)	0.284 (0.23-0.35)	0.195 (0.12-0.23)	0.144 (0.09-0.16)		0.047
WEn(s8)	0.485 (0.45-0.53)	0.479 (0.40-0.50)	0.391 (0.36-0.42)		0.010	0.033
*sign_mdetr* , raw data with mean value subtracted and mean trend removed
CWTentro	0.040 (0.03-0.05)	0.018 (0.01-0.02)	0.012 (0.011-0.014)	0.04	0.0480
CWTentro3	0.035 (0.03-0.04)	0.014 (0.009-0.019)	0.009 (0.007-0.011)		0.0480
CWTentro4	0.029 (0.02-0.03)	0.010 (0.006-0.011)	0.007 (0.006-0.011)		0.0480
WEn(s4)	0.228 (0.19-0.31)	0.159 (0.12-0.21)	0.105 (0.08-0.15)		0.0480
WEn(s5)	0.284 (0.23-0.35)	0.195 (0.12-0.23)	0.144 (0.09-0.16)		0.0480
WEn(s6)	0.389 (0.33-0.41)	0.232 (0.18-0.26)	0.197 (0.18-0.21)		0.0303
WEn(s8)	0.494 (0.45-0.53)	0.476 (0.39-0.51)	0.383 (0.35-0.42)		0.0101	0.0330
sumEn	4.599 (3.97-5.96)	6.016 (4.62-11.43)	3.683 (2.37-4.60)			0.0136
SampEn	0.007 (0.005-0.009)	0.008 (0.006-0.015)	0.005 (0.004-0.006)			0.0431

The table demonstrates the statistically significant differences of features (P refers to rank-sum P value), reflecting temperature complexity within different scales and frequency bands, between the three groups of patients. Data are presented as median with intraquartile range (25^th to 75th percentiles). CWTentro, entropy derived after continuous wavelet transformation of the whole signal; Wen, wavelet entropy; sumEn, multiscale entropy; SampEn, sample entropy; s, scale.

ISSN: 1364-8535