Skip to content

jiveshkalra/vital-sense-ai

Repository files navigation

VitalSenseAI

VitalSenseAI Banner

AI-Powered Lung and Heart Disease Detection


📚 Table of Contents


🚀 What is VitalSenseAI?

VitalSenseAI is a cutting-edge web application that leverages advanced AI models to analyze lung and heart sounds for rapid, accurate disease detection. Designed for both healthcare professionals and individuals, it empowers users to upload or record audio samples and receive instant, AI-driven health insights.


📸 Screenshots

Audio Upload Page Screenshot

Figure 1: The audio upload page where users can submit lung or heart sound recordings for analysis.

Healthy Lungs Analysis Result Screenshot

Figure 2: Example of an analysis result indicating healthy lung sounds.

COPD Detection Result Screenshot

Figure 3: AI model output showing detection of Chronic Obstructive Pulmonary Disease (COPD).

Bronchiectasis Detection Result Screenshot

Figure 4: Analysis result page displaying a diagnosis of bronchiectasis.

URTI Detection Result Screenshot

Figure 5: Example result for an Upper Respiratory Tract Infection (URTI) detection.

🧠 How It Works

  1. Record: Capture lung or heart sounds using an electronic stethoscope.
  2. Upload: Securely upload the audio sample to VitalSenseAI.
  3. Analyze: My AI model, hosted on Hugging Face Spaces, processes the sound and predicts possible diseases.
  4. Diagnose: Instantly receive a detailed health report, including disease information, severity, and prevention tips.

🤖 About the AI Model

  • Model: Custom CNN based deep learning model for lung/heart sound classification
  • Github: Github Repo
  • Hosted on: Hugging Face Spaces
  • API Integration: Uses Gradio Client for seamless communication
  • Supported Diseases: Healthy, COPD, Pneumonia, Bronchiectasis, URTI, and more

🛠️ Built With


🛠️ Getting Started

Prerequisites

Installation

# Clone the repository
git clone https://github.com/jiveshkalra/vital-sense-ai.git
cd vital-sense-ai

# Install dependencies
npm install 

Running Locally

npm run dev 

Visit http://localhost:3000 in your browser.


🧪 Features

  • 🎤 Audio Upload & Recording: Upload your own lung/heart sound or try sample files
  • Instant AI Analysis: Get results in seconds
  • 📊 Detailed Results: Disease info, severity, and prevention tips
  • 🔒 Privacy First: No audio is stored; all analysis is real-time
  • 📱 Responsive: Works on mobile, tablet, and desktop

🤝 Contributing

We welcome contributions! Please see CONTRIBUTE.md for guidelines.


📄 License

This project is licensed under the MIT License. See LICENSE.md for details.


Built With ❤️ by Jivesh Kalra

About

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published