Optimized U-Net Segmentation and Hybrid Res-Net for Brain Tumor MRI Images Classification

A brain tumor is a portion of uneven cells, need to be detected earlier for treatment. Magnetic Resonance Imaging (MRI) routinely utilized procedure take images. Manual segmentation crucial task and laborious. There an automated system classification surgery medical treatments. This work suggests efficient based on deep learning techniques. Initially, Squirrel search optimized bidirectional ConvLSTM U-net with attention gate proposed tumour segmentation. Then, the Hybrid Deep ResNet Inception Model used classification. optimizer mimics searching behavior southern flying squirrels their well-organized way movement. Here, squirrel tune hyperparameters model. In addition, modules position channel were added in U-Net extract more characteristic features. Implementation results BraTS 2018 datasets show that outperforms terms accuracy, dice score, precision rate, recall Hausdorff Distance.