Brain Tumor Detection and Classification Using Deep Learning and Sine-Cosine Fitness Grey Wolf Optimization

View full-text article in PMC

. 2022 Dec 22;10(1):18. doi: 10.3390/bioengineering10010018

Algorithm 1 ADSCFGWO algorithm

1:
Initialize ADSCFGWO population $X_{i} (i = 1, 2, \dots, n)$ with size n, iterations $t_{M a x}$ , fitness function $F_{n}$ , parameters (a, $A_{1}$ , $A_{2}$ , $A_{3}$ , $C_{1}$ , $C_{2}$ , $r_{1}$ , $r_{2}$ , $r_{3}$ , $r_{4}$ )
2:
Calculate fitness function $F_{n}$ for each $X_{i}$
3:
Find best solutions as $S_{α}, S_{β}, S_{δ}$
4:
Set t = 1
5:
while $t \leq t_{M a x}$ do
6:
Update $r_{1}$ by $r_{1} = a (1 - \frac{t}{M a x_{i t e r}})$
7:
for ( $i = 1 : i \leq n_{1}$ ) do
8:
DynamicSearch( $F_{n}$ )
9:
Update Fitness by Equation (3)
10:
Update positions from GWO as $X (t + 1) = \frac{T_{1} + T_{2} + T_{3}}{3}$
11:
if ( $r_{4} < 0.5$ ) then
12:
Update positions from SCA as $X (t + 1) = X (t) + r_{1} \times sin (r_{2}) \times | r_{3} S_{α} - X (t) |$
13:
end if
14:
end for
15:
for ( $i = 1 : i \leq n_{2}$ ) do
16:
DynamicSearch( $F_{n}$ )
17:
Update Fitness by Equation (3)
18:
Update positions from GWO as $X (t + 1) = \frac{T_{1} + T_{2} + T_{3}}{3}$
19:
if ( $r_{4} \geq 0.5$ ) then
20:
Update positions from SCA as $X (t + 1) = X (t) + r_{1} \times cos (r_{2}) \times | r_{3} S_{α} - X (t) |$
21:
end if
22:
end for
23:
Update fitness function $F_{n}$ for each $X_{i}$
24:
Update parameters
25:
Find best solutions as $S_{α}, S_{β}, S_{δ}$
26:
end while
27:
Return best solution $X^{*}$
28:
DynamicSearch( $F_{n}$ )
29:
if (Best $F_{n}$ is same for three iterations) then
30:
Increase exploration group solutions $(n_{1})$
31:
Decrease exploitation group solutions $(n_{2})$
33:
end if