Early prediction of MODS interventions in the intensive care unit using machine learning

Journal of Big Data

Table 2 Logistic regression analysis of factors affecting MODS occurrence

Features	β	S.E	Wald	P	OR	95%CI
currentHour	0.000	0.000	299.352	< 0.001	1.000	0.995–1.005
currentMaxHr	0.012	0.000	5256.416	< 0.001	1.009	1.006–1.012
currentMaxDopamine	− 0.002	0.000	19.395	< 0.001	1.308	1.305–1.311
currentMinOi	0.000	0.000	52.850	< 0.001	0.881	0.878–0.884
currentMinGcs	− 0.197	0.006	1178.851	< 0.001	0.609	0.606–0.612
currentMaxLactate	− 0.018	0.002	93.723	< 0.001	1.208	1.204–1.212
currentMaxCreatinine	0.002	0.000	395.423	< 0.001	1.140	1.135–1.145
currentMaxBilirubin	0.000	0.000	1.353	0.245	1.000	0.996–1.004
currentMinPlatelet	0.000	0.000	0.788	0.375	1.001	0.994–1.008
currentCardiovascularMods	0.638	0.016	1589.776	< 0.001	1.892	1.834–1.953
currentRespiratoryMods	− 0.060	0.005	166.921	< 0.001	1.160	1.163–1.157
currentRenalMods	0.243	0.010	577.973	< 0.001	1.275	1.250–1.300
currentGcsMods	0.826	0.012	4984.566	< 0.001	1.284	1.233–1.337
currentHepaticMods	− 0.024	0.003	67.270	< 0.001	1.180	1.177–1.183
currentHematologicMods	0.334	0.010	1231.870	< 0.001	1.396	1.371–1.423
cardiovascular24HoursMods	0.801	0.005	21322.022	< 0.001	1.728	1.705–1.752
respiratory24HoursMods	0.668	0.003	37779.847	< 0.001	1.851	1.838–1.864
renal24HoursMods	1.290	0.006	42944.155	< 0.001	2.632	2.588–2.676
gcs24HoursMods	0.825	0.003	69569.494	< 0.001	3.281	3.267–3.295
hepatic24HoursMods	1.008	0.004	59456.583	< 0.001	1.941	1.919–1.963
hematologic24HoursMods	1.179	0.011	12531.788	< 0.001	1.550	1.583–1.517