https://github.com/tictechtoee/diagnosis
AI Powered Web App to Detect Pneumonia, Lung Cancer and COVID.
https://github.com/tictechtoee/diagnosis
ai django machine-learning python
Last synced: 5 months ago
JSON representation
AI Powered Web App to Detect Pneumonia, Lung Cancer and COVID.
- Host: GitHub
- URL: https://github.com/tictechtoee/diagnosis
- Owner: ticTechtoee
- Created: 2024-06-03T18:48:51.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2024-12-17T10:50:35.000Z (over 1 year ago)
- Last Synced: 2025-08-27T10:39:07.300Z (10 months ago)
- Topics: ai, django, machine-learning, python
- Language: CSS
- Homepage: https://portfoliofarhan.pythonanywhere.com/
- Size: 22 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# **AI-Powered Medical Diagnostic Tool** 🩺
**Detect Pneumonia, COVID-19, and Lung Cancer using Machine Learning and Medical Imaging**
## 🚀 **Overview**
This project implements an AI-driven diagnostic application capable of detecting:
- **Pneumonia** and **COVID-19** from Chest X-Rays
- **Lung Cancer** from Chest CT Scans
The application combines a **Convolutional Neural Network (CNN)** for image classification with a web interface built using Django, providing an accessible and user-friendly diagnostic solution.
---
## 🛠️ **Technologies Used**
### **Machine Learning**
- **TensorFlow/Keras**: Deep learning framework for model building and training
- **NumPy**: Efficient array manipulation and computations
### **Web Development**
- **Django**: Python-based backend web framework
- **Bootstrap**: Front-end framework for responsive UI design
### **Data Preprocessing**
- **OpenCV**: Image processing for grayscale conversion, resizing, and enhancements
- **Sklearn**: For data splitting (train-test split)
---
## 🧪 **How It Works**
### **1. Data Preprocessing**
The `preprocess.py` script:
- Loads X-Ray or CT Scan images from the dataset directory.
- Converts images to grayscale and resizes them to **224x224 pixels**.
- Normalizes pixel values to the range [0, 1].
- Splits the data into training and testing sets.
```bash
# Run preprocessing to prepare the dataset
python preprocess.py
```
### **2. Model Training**
The `build_model.py` script trains a **Convolutional Neural Network (CNN)** for multi-class classification:
- 3 Classes for Pneumonia/COVID Detection: `COVID`, `Non-COVID`, and `Pneumonia`
- 2 Classes for Lung Cancer Detection: `Lung Cancer` and `Normal`
Training features include:
- **3 Convolutional Layers** with ReLU activation
- **MaxPooling** for dimensionality reduction
- **Dropout** to avoid overfitting
- **Softmax** activation for multi-class outputs
```bash
# Train the machine learning model
python build_model.py
```
### **3. Disease Detection via Web Application**
The `detect.py` file integrates the trained models into a Django web application.
- Upload chest X-Ray/CT images through the frontend.
- Process the uploaded image and predict results using the model.
- Return diagnosis (e.g., COVID, Pneumonia, or Lung Cancer) with status updates.
**Sample Workflow**:
1. Upload an image (`.jpg` or `.png`)
2. Process the image with the selected ML model.
3. Display prediction results on the UI.
---
## 🌟 **Features**
- **Multiple Disease Detection**: Predict COVID-19, Pneumonia, and Lung Cancer.
- **Responsive Web UI**: Built with Django and Bootstrap for usability.
- **Fast and Accurate Predictions**: Achieved with optimized CNN models.
- **Image Preprocessing**: Automatic image resizing and normalization for consistent results.
---
## 🖥️ **Folder Structure**
```plaintext
|-- assets/
| |-- models/ # Pre-trained models (.h5 files)
|-- dataset/ # Dataset of X-Ray and CT Scan images
|-- templates/ # HTML templates for Django
|-- static/ # Static files (CSS, JS, images)
|-- build_model.py # ML model training script
|-- preprocess.py # Image preprocessing script
|-- detect.py # Disease detection logic (Django view)
|-- requirements.txt # Project dependencies
|-- manage.py # Django app manager
|-- README.md # Project documentation
```
---
## ⚙️ **Setup Instructions**
1. **Clone the Repository**
```bash
git clone https://github.com/ticTechtoee/diagnosis
cd medical-diagnostic-tool
```
2. **Create Virtual Environment**
```bash
python -m venv venv
source venv/bin/activate # For Linux/Mac
venv\Scripts\activate # For Windows
```
3. **Install Dependencies**
```bash
pip install -r requirements.txt
```
4. **Prepare Dataset**
Place your dataset (Chest X-Ray/CT images) in the `dataset/` folder, organized into subfolders:
```plaintext
dataset/
|-- COVID/
|-- Non-COVID/
|-- Pneumonia/
|-- Lung_cancer/
|-- Normal/
```
Run preprocessing:
```bash
python preprocess.py
```
5. **Train the Model**
```bash
python build_model.py
```
6. **Run the Web Application**
```bash
python manage.py runserver
```
Visit `http://127.0.0.1:8000/` in your browser to use the application.
---
## 📊 **Results**
The CNN models achieved high accuracy on test datasets:
- **Pneumonia/COVID Detection**: Accurate classification into `COVID`, `Non-COVID`, and `Pneumonia`.
- **Lung Cancer Detection**: Effective binary classification into `Lung Cancer` or `Normal`.
---
## 👨💻 **Future Improvements**
- Add real-time image analysis using webcams.
- Improve accuracy with larger datasets and transfer learning.
- Deploy the application on cloud platforms for global access.
---
## 🤝 **Contributing**
Contributions are welcome! If you'd like to improve this project, feel free to fork the repository, create a new branch, and submit a pull request.
---
## 📝 **License**
This project is licensed under the MIT License.
---
## 📫 **Contact**
For collaborations, inquiries, or feedback:
- **Name**: Farhan Basheer
- **LinkedIn**: [Farhan's LinkedIn](https://www.linkedin.com/in/farhanbasheerofficial10)
- **Email**: farhanbasheerofficial10@gmail.com
---
### **If you find this project useful, don’t forget to ⭐ star the repository!** 🌟