A Hybrid DE-RGSO-ELM for Brain Tumor Tissue Categorization in 3D Magnetic Resonance Images

Algorithm 1

Refined GSO algorithm.

Set 𝑘:= 0;

Randomly initialize positions 𝑋_𝑖 and head angles 𝜙 of all members;

WHILE (the termination conditions are not met)

 For (each members 𝑖 in the group)

    Calculate fitness:    Calculate the fitness value of current member: 𝑓(𝑋_𝑖)

    Choose producer:  Find the producer 𝑋_𝑝 of the group;

    Perform producing: (1) The producer will scan at zero degree and then scan laterally by randomly sampling three points

                 in the scanning field using (10) to (12).

             (2) Find the best point with the best resource (fitness value). If the best point has a better resource

               than its current position, then it will fly to this point. Otherwise it will stay in its current

               position and turn its head to new angle using (13).

             (3) If the producer cannot find a better area after 𝛼 iterations, it will turn its head back to zero

               degree using (14);

    Perform scrounging: Randomly select 75% from the rest members to perform scrounging using (15);

    Perform ranging:    For the rest 25% members, they will perform ranging:

               Find the worst point with the worst resource (fitness value). If the worst point has a better

               resource than its current position, then it will fly to this point. These members will update

               their position based on this worst resource.

 END FOR

 Set 𝑘:= 𝑘+1;

END WHILE