Prediction of COVID-19 using Chest X-ray Images (Convolutional Neural Network - CNN) 🦠🔍📊

This project focuses on utilizing Convolutional Neural Networks (CNNs) to predict whether patients are infected with COVID-19 based on their chest X-ray images. By leveraging deep learning techniques, we aim to contribute to early detection and diagnosis of COVID-19, a critical need during the ongoing pandemic.

📌 Project Overview

In this project, we accomplished the following:

• Built and trained a CNN from scratch using Keras with TensorFlow as the backend.
• Utilized Matplotlib for data visualization to gain insights from the X-ray images.
• Performed data preprocessing and augmentation using TensorFlow 2.0 to enhance model performance.
• Designed a sequential model architecture, incorporating convolutional layers, pooling layers, dropout layers, dense layers with ReLU activation, and an output layer with sigmoid activation.
• Employed a labeled dataset containing chest X-ray images of both non-COVID and COVID-19 infected patients, obtained from Kaggle.
• Evaluated the model using various metrics, including training set accuracy, validation set accuracy, and test set accuracy.
• Employed the Adam optimizer with a learning rate of 0.001 for efficient gradient descent.
• Conducted the entire project within the Google Colab environment for seamless development and execution.

📈 Model Performance

The CNN model yielded the following impressive results:

• Training Set Accuracy: 97.38%

• Training Set Loss: 0.0875

• Validation Set Accuracy: 97.79%

• Validation Set Loss: 0.0744

• Test Set Accuracy: 98.14%

• Test Set Loss: 0.06641

🛠️ How to Use

1. Clone this repository to your local machine.
2. Open and run the project in Google Colab or your preferred development environment.
3. Explore the code, data preprocessing, model architecture, and evaluation metrics.

💡 Future Enhancements

This project serves as a foundation for further research and development in the field of medical image analysis. Potential enhancements include fine-tuning the model architecture, exploring transfer learning techniques, and expanding the dataset for even more accurate predictions.

👥 Contributors

This project was made possible by the collaborative efforts of:

• [Sainik Khaddar]

📞 Contact

For inquiries or collaboration opportunities, please feel free to reach out to us:

Email: sainikwarror132@gmail.com

📝 License

This project is licensed under the MIT License.

Google Colab Project Link : https://colab.research.google.com/drive/1oeA2Rp5B_8VWtOXEoY7R4dQMaO4pZe_y?usp=sharing

---