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

Table 11

Comparison result of ELM and IELM-PSO on SPL analysis.

Model	GA-ELM [33]	GA-IELM-PSO [28]	Proposed DE-ELM-RGSO
Norm	432.56	185.34	120.96
Condition	24.825𝑒+5	4.665𝑒+5	3.366𝑒+5
MSE error	0.966	0.991	0.932
Training efficiency %
Mean	92.92	97.42	98.26
STD	6.05	5.92	6.25
Testing efficiency %
Mean	91.27	96.85	97.12
STD	6.89	5.45	5.76
Hidden neurons	27	5	5
Accuracy %	92.8	97.14	98.29
Time
Min.	3	3.5	3