Brain Tumor Classification Using Machine Learning

Project Overview

This project focuses on the classification of brain tumors using image processing and machine learning techniques. The dataset consists of MRI images categorized into two classes: "No Tumor" and "Pituitary Tumor". The goal is to develop a model that can accurately classify the images into these categories.

Project Structure

• Training Data: Located in the Training directory, containing subdirectories for each class (no_tumor and pituitary_tumor).
• Testing Data: Located in the Testing directory, used to evaluate the performance of the model.

Libraries and Tools

The following Python libraries are used in this project:

• numpy: For numerical operations and handling arrays.
• pandas: For data manipulation and analysis.
• matplotlib: For data visualization.
• scikit-learn: For machine learning models and evaluation.
• opencv (cv2): For image processing.

Steps in the Project

1. Data Loading and Preprocessing:

   • Images are loaded from the Training directory.
   • Each image is resized to 200x200 pixels and converted to grayscale.
   • The images are flattened into 1D arrays and stored in X.
   • Corresponding labels are stored in Y, where 0 indicates "No Tumor" and 1 indicates "Pituitary Tumor".

2. Data Splitting:

   • The dataset is split into training and testing sets using an 80-20 split.

3. Feature Scaling:

   • Pixel values are scaled to a range of [0, 1] by dividing by 255.

4. Model Training:

   • Two machine learning models are trained: Logistic Regression and Support Vector Classifier (SVC).
   • The models are trained on the training set (xtrain, ytrain).

5. Model Evaluation:

   • The accuracy of both models is evaluated on the training and testing sets.
   • The SVC model is further used for prediction and visualization.

6. Misclassification Analysis:

   • Misclassified samples are identified and printed.

7. Visualization of Predictions:

   • Predictions on new test images are visualized with the predicted class labels.

Results

• The training and testing accuracy for both models are printed.
• The total number of misclassified samples is displayed.
• Sample predictions are visualized using the SVC model, with images from both classes ("No Tumor" and "Pituitary Tumor").

How to Run

1. Ensure all required libraries are installed:

   pip install numpy pandas matplotlib scikit-learn opencv-python