Paper Title
Brain Tumor Detection and Classification Using Deep Learning

Abstract
Due to the difficulty of brain segmentation, this article presents a strategy for removing brain tumors from 3D magnetic resonance images (MRI) and computed tomography (CT) scans using ResNet50 and the 3D U-Net design adopted by the deployment strategy. ResNet50 achieves 98.97% accuracy in this query, while 3D U-Net achieves 98.99% accuracy in many deep learning methods. It is worth mentioning that the traditional convolutional neural network (CNN) provides 98.It is worth noting that the Traditional Convolutional Neural Network (CNN) provides 98.90% accuracy in 3D MRI. In addition, the image fusion method combines multiple images to create a merged image to remove highlights from the medical image. In addition, we determine the loss function using several indicators and get the dice results on a particular indicator. The average score for Dice Coefficient and Soft Dice Loss for our tests is 0.0990. Meanwhile, the expected and recorded values for these two experiments using the region-level estimates are 0.0311 and 0.5887, respectively. On the other hand, an integrated set of channels is put together to send the central model to the web server so that it can be accessed by software systems using the Agent REST API. Finally, the proposed model was validated with the Area-Under-Curve- Acceptor Characteristic Operator (AUC-ROC) curve and confusion matrix and compared with the current research paper for problem capacity. Through comparative analysis, we extracted brain segmentation concepts and identified differences. However, the proposed model can be adapted in everyday life and therapy to identify areas of disease and tumor in the brain with various measurements. Keywords - Brain tumor Detection ;Deep Learning; Deep Neural Network MRI; CNN Profile.; Python