. Author manuscript; available in PMC: 2017 Jul 27.

Published in final edited form as: Physiol Meas. 2017 May 10;38(6):1061–1076. doi: 10.1088/1361-6579/aa6e9c

Table 4.

Features extracted from quiescent segments of of RR intervals, shown as medians and IQR bounds in brackets. CTRL refers to the control group. Test AUC reports performance of univariate classifier trained solely on one feature.

	PTSD status

Feature	PTSD	CTRL	Test AUC
AC (s)^a,^b	−9.62 [−1.26e1 −6.22]	−1.28e1 [−1.91e1 −9.72]	0.77 [0.73 0.82]
DC (s)^a,^b	9.43 [6.64 1.22e1]	1.40e1 [1.11e1 2.06e1]	0.82 [0.73 0.84]
LF power (s²)^a,^b	3.31e2 [1.52e2 5.78e2]	8.71e2 [4.44e2 1.47e3]	0.81 [0.75 0.88]
σ_rr (s)^a	4.14e−2 [3.44e−2 5.34e−2]	7.12e−2 [4.9e−2 8.06e−2]	0.82 [0.73 0.84]
IQR_rr (s)^a	5.40e−2 [3.55e−2 5.60e−2]	7.20e−2 [5.50e−2 9.38e−2]	0.78 [0.71 0.81]
SDNN (s)^a,^b	4.68e−1 [3.16e1 5.97e1]	6.47e1 [4.32e1 7.70e1]	0.75 [0.57 0.86]

P < 0.05 comparing feature values from PTSD versus control subjects via two-sided Kolmogorov–Smirnov test.

Feature among combination that maximizes training set AUC.