Clinical Outcome Prediction in Aneurysmal Subarachnoid Hemorrhage Using Bayesian Neural Networks with Fuzzy Logic Inferences

Table 2

Results of Bayesian regression analysis of predictors of neurologic outcome in aneurysmal subarachnoid hemorrhage using uninformed priors. Output generated by WinBUGS version 1.4.3.

Node	Mean	SD	MC error	2.5%	Median	97.5%	Start	Sample
b.MOTOR	−0.3268	0.0833	2.606𝐸−4	−0.4904	−0.3266	−0.1632	5000	95001
b.CVA	0.9499	0.04679	1.589𝐸−4	0.8582	0.95	1.041	5000	95001
b.BSWELL	0.4307	0.06552	2.262𝐸−4	0.3021	0.4309	0.5596	5000	95001
b.MI	0.2964	0.1454	4.414𝐸−4	0.01064	0.2966	0.5805	5000	95001
b.NEUROGR	0.2762	0.02807	8.489𝐸−5	0.2213	0.2763	0.3313	5000	95001
b.DM	0.254	0.1043	3.384𝐸−4	0.04876	0.2545	0.4568	5000	95001
b.ADMITVSP	0.2417	0.06948	2.481𝐸−4	0.1046	0.2418	0.3775	5000	95001
b.PREVSAH	0.1747	0.06821	2.285𝐸−4	0.04019	0.1748	0.3073	5000	95001
b.ANSIZE	0.2124	0.03838	1.166𝐸−4	0.1373	0.2123	0.288	5000	95001
b.IVH	0.1727	0.04227	1.329𝐸−4	0.08972	0.1728	0.2551	5000	95001
b.HTN	0.1223	0.04511	1.45𝐸−4	0.03416	0.1223	0.2108	5000	95001
b.VSPDAY	0.05136	0.02759	8.475𝐸−5	−0.002518	0.05136	0.1055	5000	95001
b.AGE	0.01581	0.001591	5.454𝐸−6	0.0127	0.01581	0.01895	5000	95001
b.MAP	0.004209	0.001252	4.197𝐸−6	0.001761	0.004204	0.006663	5000	95001
b.D8TEMP	−0.08723	0.04131	1.333𝐸−4	−0.168	−0.08717	−0.006547	5000	95001