A healthcare chatbot that provides medical diagnosis and recommendations based on user input symptoms using a fine-tuned model deployed on Hugging Face.
Check out the deployed Model on Hugging Face.
If you are using the deployed app for testing then kindly wait for 40 sec - 50 sec after the first input. It takes around 50sec to restart the server again.
This project involves a chatbot that accepts symptoms from the user and returns possible medical conditions and recommendations. It leverages a fine-tuned BERT model deployed on Hugging Face for the predictions.
- Chat interface for entering symptoms.
- Real-time medical diagnosis and recommendations.
- Typing effect for bot responses.
- React
- Axios
- Typing Effect Hook
- CSS
- Node.js
- Express.js
- Axios
- Hugging Face API
- Node.js and npm installed on your machine.
-
Clone the repository:
git clone https://github.com/RishiRaj0712/healthcare_chatbot.git cd healthcare_chatbot
-
Install frontend dependencies:
cd diagnosis-chatbot npm install
-
Install backend dependencies:
cd ../diagnosis-chatbot-backend npm install
Create a .env file in the diagnosis-chatbot-backend directory and add your Hugging Face API token:
HUGGING_FACE_API_TOKEN=your_hugging_face_api_token
-
Navigate to the backend directory:
cd diagnosis-chatbot-backend
-
Start the backend server:
node index.js
-
Open a new terminal and navigate to the frontend directory:
cd diagnosis-chatbot
-
Start the frontend development server:
npm run dev
-
Open your browser and navigate to http://localhost:5173 to interact with the chatbot.
This project is licensed under the MIT License.