🫁 Lung Disease Detection Using Audio Classification

This is a web-based application that detects various lung diseases from audio recordings of respiratory sounds using deep learning and audio signal processing.

🔍 Features

• 🎤 Upload respiratory audio files (e.g., breath or lung sounds)
• 🎼 Convert audio to spectrograms using librosa
• 🧠 Predict diseases using a Convolutional Neural Network (CNN)
• 💡 Detects conditions like:
  • Asthma
  • COPD
  • Pneumonia
  • Lung Fibrosis
  • Pleural Effusion
  • Heart Failure-related conditions
• 🌐 Built with Flask and rendered with interactive HTML/CSS/JS
• 📊 Displays diagnosis result along with the spectrogram visualization

🛠️ Tech Stack

• Frontend: HTML, CSS, JavaScript
• Backend: Python, Flask
• Deep Learning: TensorFlow/Keras
• Audio Processing: Librosa, Matplotlib
• Visualization: Spectrogram generation and display

🚀 How It Works

1. User uploads an audio file through the web interface.
2. The system extracts a spectrogram from the audio input.
3. The spectrogram is passed to a CNN model trained to classify lung diseases.
4. The prediction result and spectrogram image are displayed.

🗂️ Project Structure

lung-disease-detection/
├── app.py
├── templates/
│   ├── index.html
│   └── result.html
├── static/
│   └── [spectrogram images]
├── uploads/
│   └── [uploaded audio files]
├── requirements.txt
└── README.md

▶️ Getting Started

1. Clone the repository

git clone https://github.com/yourusername/lung-disease-detection.git
cd lung-disease-detection

2. Install dependencies

pip install -r requirements.txt

3. Run the Flask app

python app.py

4. Open in browser

Visit http://127.0.0.1:5000/ in your browser to use the application.

📄 License

This project is for educational and research purposes only.