Leveraging Pre-Trained Deep Learning Models for Brain Tumor Detection and Classification from MRI Scans

Medical imaging in the form of neuroimageries is of immense help in the detection of brain tumor and for planning their treatment. As we progress in the deep learning section, preexisting models like the VGG16, ResNet, and Inception have a lot of potentiality in classifying the brain tumor. Brain Tumors vary in size, form, and texture, brain they are difficult to identify. Due to the limitations of traditional CNN models, ResNet uses residual connections to extract deep hierarchical features, helping to preserve crucial detail. Its inception sections, which apply various filter sizes in parallel, capture multi-scale features. VGG16: Ensures fine-grained feature extraction by offering an extensive sequencing architecture with smaller receptive fields, A deep learning model with a hybrid approach that combines the advantages of ResNet, a Inception, and VGG16 to increase classification accuracy. Combine features from multiple models to improve accuracy. Classify brain tumors into different categories. Hierarchical properties are captured differently by various models. Makes use of several feature representations to lessen overfitting ResNet and VGG16 assure deep spatial feature extraction, while Inception record multi-scale features. Effectively manages various brain tumor forms and textures. Proposed model achieves the Classification accuracy 93-96%. Computationl Complexity is overcome and give the Balanced Performance.Reduced overfitting due to feature fusion, overcome the disadvantages of the Existing model and Give the Better Performance