Automated segmentation of brain tumour images using deep learning-based model VGG19 and ResNet 101

A life-threatening neurological illness known as a brain tumour is brought on by the uncontrolled growth of cells inside the brain or the skull. Brain tumours may be fatal. A life-threatening brain tumour is caused by uncontrolled cell growth in the brain or skull, which may lead to death. The massive quantity of data that an MRI generates makes it impossible to manually segment the data in a reasonable length of time, which limits the use of accurate quantitative measures in clinical practice. The success of a treatment plan for brain cancer is highly dependent on the level of training and expertise of the attending physician. As a result of this, the utilization of an automated tumour detection system is of the utmost importance in order to assist radiologists and physicians in the detection of brain tumors. Hence, Methods of reliable and automatic segmentation are necessary. However, due to the high amount of spatial and anatomical diversity that exists across brain tumours, automatic segmentation presents a difficult challenge. In this article, we proposed a hybrid model for the automatic segmentation of images that are based on the VGG 19 stack using convolutional neural networks (CNN) with ResNet101. The optimization of a classifier is performed by Bayesian and then cross-validation is used to choose the best parameters for optimized classifiers. During the course of the studies, a number of different transfer learning models were evaluated to, ultimately, decide which model is the most effective for identifying brain alignancies on the basis of neural networks. The suggested stacked classifier model, which makes use of the most recent and cutting-edge technology, outperforms all previous models with regard to precision, recall, and F1 scores.