Fig 1.
Framework block diagram of the G-Net.
Fig 2.
Squeeze Excitation module when a single image passes through.
Table 1.
Description and rationale of various blocks.
Table 2.
Purpose, integration mechanism, and benefits of G-Net.
Table 3.
Description of dataset.
Fig 3.
MRI sequences as 2D sections.
Table 4.
Performance comparison of G-Net for different epochs.
Table 5.
Performance comparisons of Subclass tumor using Brats 2020.
Table 6.
Performance comparisons of subclass tumor using Brats 2021.
Table 7.
Comparison of G-Net with existing techniques on BraTS2020.
Fig 4.
Accuracy, loss when trained for 18 epoch.
Fig 5.
Precision, sensitivity and specificity when trained for 18 epoch.
Fig 6.
Accuracy, loss when trained for 50 epoch.
Fig 7.
Precision, sensitivity and specificity when trained for 50 epoch.
Fig 8.
Performance comparison of accuracy, precision, sensitivity, specificity for 18 and 50 epochs.
Fig 9.
Performance comparison of loss, tversky loss, boundary loss for 18 and 50 epochs.
Fig 10.
Illustration presented below displays the qualitative outcomes of the G-Net model applied to BraTS2020.
Fig 11.
Following visualization showcases the results of the G-Net over existing models ([5, 10, 11, 18, 42, 44]) on BraTS2020.
Table 8.
Model integration and performance analysis of G-Net.