Brain Tumor Detection and Classification Using Intelligence Techniques An Overview

Aim:

        To detect and identify the Brain Tumor using Deep-Learning techniques

Abstract:

       Brain is the controlling unit of human body. It regulates the functions such as memory, vision, hearing, knowledge, personality, problem solving etc. The main reason for brain tumors is the uncontrolled development of brain cells. In medical practices, the early detection and recognition of brain tumors accurately is very vital. In literature, there are many techniques has been proposed by different researchers for the accurate segmentation of brain tumor. Magnetic resonance imaging (MRI) is high-quality medical imaging, particularly for brain imaging. MRI inside the human body is helpful to see the level of detail. The MRI is used even in diagnosis of most severe disease of medical science like brain tumors. The brain tumor detection process consist of image processing techniques involves four stages. Image pre-processing, image segmentation, feature extraction, and finally classification.

Proposed System

       The diagnosis of Tumor disease detection at the early stages is very important. Our proposed methodology is based on Deep Neural Network Model with various Transfer Leaning models which trains on the Dataset and detects the image with a classification and in such image the disease gets segmented.

Advantages:

      Transfer learning, a technique in deep learning, involves leveraging pre-trained models on one task to enhance performance on a different but related task. The ResNet50,Alexnet,VGG16  model is a prime example of a deep neural network that excels in transfer learning due to its versatile architecture and broad pre-training Model Architecture: Alexnet,VGG16,ResNet50  are convolutional neural network (CNN) architecture known for its efficiency in image analysis. It features various “Inception modules,” which consist of parallel convolutional operations of different sizes and pooling operations. This design enables the model to capture features at multiple scales, aiding in recognizing complex patterns within images. Pre-Training on ImageNet: Before transfer learning, undergoes pre-training on massive datasets, such as the ImageNet dataset. This phase imparts the model with a diverse set of features, including edges, textures, and higher-level object parts, learned from a wide range of images. Fine-Tuning for Task-Specific Objectives: After pre-training, fine-tuned for a specific task. Fine-tuning typically involves modifying the final layers of the network to align with the new task’s objectives. For example, if the task is medical image classification, the output layer might be changed to accommodate the number of classes in the medical dataset.