Ayoub Yousefi Jourdehi 120

madbrainedmudlickΤεχνίτη Νοημοσύνη και Ρομποτική

20 Οκτ 2013 (πριν από 3 χρόνια και 9 μήνες)

69 εμφανίσεις




General biology

Poster

Modeling of soy
dietary
g
enistein and
e
quol

effects
on
plasma phosphorous

levels

by
using artificial neural network
in

farmed
female
beluga (
Huso huso
)

Yousefi Jourdehi

A

1
, 2
*
,
S
udagar

M
1
, Bahmani

M
2
,
Hos
s
eini
S.
A
1
,
Dehghani A.
A
1
,
Yazdani

M.A

2

1
-

Gorgan
University of
Agricultural
Sciences &
Natural
Resource
s

2

-

International Sturgeon Research Organization of the Caspian Sea

Ayoub2222002@yahoo.com

Abstract

An artificial neural network
(ANN) uses a highly interconnected group of simulated neurons
(units, elements, or nodes) that process information in parallel.

In this study,
The

data

extracted from

54
Huso huso

5


year


old
were
fed by diets containing different
concentrations
(
0.2,
0.4, 0.8, 1.6 gr/kg
)

of
soy
-

phytoestrogen
s

genistein and equol

during a
year,
were used to train multi
-
layer

feed
-
forward artificial neural networks.
The input
parameters are

phytoestrogens concentrations

and seasons

and
the output parameter
is
phosphorus level.
Various applied networks easily generated

associations

of
plasma
phosphorous

depends on

gonad

development
,

providing a powerful tool

for
estimation
. The
accuracy

of the trained network was tested with data from

fish
.
Based on results,
n
eural
n
etwork system predicted phosphorus levels with high performance r

= 0.99.
Therefore,
ANN
can
o
ff
er

as

an alternative
approach
when there are significant

overlaps in
phosphors levels.


Performance

Y3

MSE

1.389786844

r

0.996586805





Key

words
:
A
rtificial neural network
,

Phytostrogens,
Huso huso