https://github.com/ali-jaan-butt/brain_tumor_classification

Brain Tumor Detection using Deep Learning 🧠
https://github.com/ali-jaan-butt/brain_tumor_classification

brain-tumor-classification deep-learning images-classification opencv

Last synced: 3 months ago
JSON representation

Brain Tumor Detection using Deep Learning 🧠

• Host: GitHub
• URL: https://github.com/ali-jaan-butt/brain_tumor_classification
• Owner: Ali-Jaan-Butt
• License: mit
• Created: 2025-08-19T06:43:28.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2025-08-19T07:34:00.000Z (11 months ago)
• Last Synced: 2025-09-13T21:58:34.303Z (10 months ago)
• Topics: brain-tumor-classification, deep-learning, images-classification, opencv
• Language: Jupyter Notebook
• Homepage:
• Size: 7.81 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md
  • License: LICENSE
  • Security: SECURITY.md

Awesome Lists containing this project

README

          
# Brain Tumor Detection using Deep Learning 🧠

This project focuses on detecting **brain tumors from MRI images** using **Convolutional Neural Networks (CNNs)** with **transfer learning (ResNet50)**. The notebook includes preprocessing, model training, and evaluation steps to classify MRI scans as **tumor** or **non-tumor**.

---

## 📂 Dataset

The dataset used is [Brain MRI Images for Brain Tumor Detection](https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection) available on Kaggle.  

- Images are categorized into **two classes**:  

  - `Yes` → MRI showing brain tumor  

  - `No` → MRI without tumor  

---

## ⚙️ Workflow

The notebook follows these main steps:

1. **Import Libraries**  

   Uses `numpy`, `pandas`, `matplotlib`, `seaborn`, `opencv`, `tensorflow/keras`.

2. **Data Preprocessing**  

   - Load images and labels  

   - Split into train and test sets (80/20)  

   - Image normalization and augmentation

3. **Model Architecture**  

   - Based on **ResNet50** with additional Dense and Dropout layers  

   - Activation: ReLU, Softmax  

   - Optimizer: Adam  

4. **Training**  

   - Early stopping used to prevent overfitting  

   - Data augmentation applied via `ImageDataGenerator`  

5. **Evaluation**  

   - Accuracy, Confusion Matrix, and Classification Report  

---

## 📊 Results

- Achieved **high classification accuracy** on test data  

- Clear distinction between tumor and non-tumor cases  

- Visualized performance using plots and confusion matrix  

---

## 🚀 Requirements

Make sure you have the following installed:

```bash

pip install numpy pandas matplotlib seaborn opencv-python tensorflow pillow scikit-learn