Brain Tumor Detection

This project uses a deep learning model for brain tumor detection based on magnetic resonance imaging (MRI). The notebook Brain_Tumor_Detection.ipynb contains all the code necessary to train and evaluate the model.

Project Overview

This project consists of three main parts:

1. Data Preprocessing:

   • Splitting the Data: The images are divided into training, validation, and test sets.
   • Cropping: A crop is applied to the images.
   • Resizing: The images are resized to a standard dimension.
   • Data Augmentation: Various data augmentation techniques are applied to increase the diversity of the dataset.

2. Neural Network Construction:

   • Transfer Learning: The neural network is built using transfer learning, specifically employing the ResNet50 architecture.

3. Testing Phase:

   • Model Testing: The trained model is tested to evaluate its performance.

Requirements

• Python 3.x
• Jupyter Notebook
• TensorFlow
• Keras
• NumPy
• Matplotlib
• Scikit-learn
• OpenCV

Architecture

The architecture of the network includes:

1. Input Layer: Accepts images resized to 224x224 pixels.
2. Pre-trained Base Model: The ResNet50 model pre-trained on the ImageNet dataset is used as the base. This model includes:
   • Convolutional layers that extract features from the input images.
   • Batch normalization and activation layers to stabilize and enhance training.
   • Residual blocks to allow for deeper networks by alleviating the vanishing gradient problem.
3. Global Average Pooling Layer: Reduces the spatial dimensions of the feature maps from the base model, providing a 1D feature vector.
4. Dropout Layer: A dropout layer with a dropout rate of 0.2 to prevent overfitting.
5. Output Layer: A dense layer to output the prediction.