KAICare

The very first AI-powered physical therapy platform that helps you recover from injuries, improve your performance, and prevent future injuries.

KAICare is a compassionate haven where cutting-edge technology and empathetic care converge. Led by our AI companion, KAI the koala, we empower patients on their healing journey.

Features:

• KAI (AI Assistant) - KAI is an AI-powered assistant that helps you with your physical recovery when you need it the most. KAI will guide you through your recovery process, provide you with personalized exercises, and track your progress.
• Personal Trainer - Computer Vision-aided agent that corrects your pose while doing your recovery exercises.

About the team

This application has been developed by 5 Illinois Institute of Technology engineer graduate students for the Orahacks hackathon that took place in April 2024.

• Si Qiang Wu Yang | LinkedIn | [email protected] | Computer Vision Developer & Design
• Chenglong (Oskar) Wu Yang | LinkedIn | [email protected] | Front/Web3 Developer & Design
• Mario de Lucas Garcia | LinkedIn | [email protected] | LLM/Front Developer & Design
• Ignacio Gomez Valverde LinkedIn | [email protected] | DB/LLM Developer & Design
• Laura Serrano Velazquez | LinkedIn | [email protected] | Front Developer & Design

Technologies

• Tailwind: Design, animations and effects
• Clerk: User Authentication with Google, Apple & MetaMask
• Web3: Use of authentication through MetaMask
• Atlas MongoDB: LLM knowledge storage & vector similarity search
• AWS: Host for the MongoDB cluster
• BRX.AI: LLM deployment
• AI: LLM & Computer Vision
• AI Framework: TensorFlow

Web navigation

The following figure shows the flow possibilities currently configured in KAICare Web

Set Up & Deployment

To run the program, the following steps must be performed.

1. Clone the repository

git clone https://github.com/KaiCare/web.git 
cd web

2. Install the necessary packages - via npm package manager

npm install

If the previous command gives an error, it is probably because you do not have npm installed.

3. Setup .env file

To facilitate the creation of the .env file with the necessary data to run KAICare we have created a configuration.py file that checks if the .env file is already created and if not it creates it with all the necessary information inside.

python3 configuration.py

4. Start Computer Vision model

cd server
pip install tensorflow==2.4.1 tensorflow-gpu==2.4.1 opencv-python mediapipe scikit-learn matplotlib testresources
python3 Exercise_detector.py

5. Deploy the app (in another terminal) - http://localhost:3000

npm run dev