Abstract
In the past few years, Brain Tumors (BTs) have caused high mortality and morbidity in the global population. To ensure the life cycle of BT individuals, an automatic and precise framework for BT detection and survival rate estimation is essential. However, the traditional methods failed to concentrate on the necrotic core in BT classification. Hence, a promising framework called SM-GN-RNN and Cr-FLS-based BT classification and survival rate prediction is proposed in this paper. Firstly, various individuals' brain fMRI is collected and then pre-processed. Next, the active contour technique is utilized in segmenting the tumor from the pre-processed images. After that, the tumor volume is calculated from the segmented images. Similarly, the necrotic area is identified from the segmented images by using tile generation and mapping. Subsequently, the feature extraction is done; following that, the optimal features are selected using the Gm-LOA. Afterward, the optimal features are inputted to the SM-GN-RNN, which classifies the benign and malignant tumor. Lastly, the malignant tumor and tumor volume are subjected to Cr-FLS, which predicts the survival rate of the BT individuals. Thus, the experimental findings stated that the proposed methodology is the eye-opening framework for reducing the mortality rate due to BT .
