This is a Neural Network based prediction of long-term outcome in Type A aortic dissection patients.

		Min Exp	Max Exp
Age		27	85
Annint		2002	2008
Arrc		0	200
Bl24		0	6710
Cint		0	1
Cren		0	1
Cvvh		0	1
dDN0		0	1
dDN1		0	1
dDN2		0	1
HP		0	1
dInt1		0	1
dInt2		0	1
dInt3		0	1
dInt4		0	1
dInt5		0	1
dInt6		0	1
IRC		0	1
IscL		0	1
dPerBr1		0	1
dPerBr2		0	1
dPerBr3		0	1
HOSP		0	1
PoDN		0	1
PoIp		0	2
PoMD		0	1
PoSk		0	1
PoTOT		0	1
Redo		0	1
Sex		0	1
TCEC		24	682
VAB		0	1
Prediction
Status		Strength =

'Max/Min Exp' - the maximum and minimum values experienced during the creation of the model. Visual warnings are given if your input values are outside this range of experience.
'Strength' - this is based on a scale from -1 to +1, where 0 is a borderline case and +1 is the best TRUE case and -1 is the best FALSE case on the training data. The higher the absolute value of the number, the stronger the classification.

---

inputs:	32
hidden neurons:	2
training data source:	G:\DATI-PENDR\Mac1\DIS2.xls
training data view:	Foglio1$
Tiberius weights file:	G:\PAPERSandSCIENCE\Papers0910\EurJCThSurg\Model2Chronic.twf
model created:	ven, nov 27 2009, 2.59
this page created:	lun, gen 11 2010, 11.42

---

Visit the Tiberius Web Site