Dispatch Simulator
Dispatch Simulator Grading
Login signup page
Home page

WatchOver: AI-Powered Emergency Training Platform

Problem Statement

The lack of effective training tools for emergency dispatchers and responders often leads to delayed response times and suboptimal decisions during critical situations. This gap is particularly significant in addressing challenges such as scalable dispatcher training, enhancing real-time decision-making, and improving collaboration between emergency teams. These issues align with the Laerdal challenges of creating life-saving solutions through technology.

Solution: WatchOver

WatchOver addresses these challenges by providing:

AI-driven simulations for realistic dispatcher training.
A grading and feedback system to evaluate decision-making skills.
A seamless integration of first responders into emergencies using real-time communication.

The platform is designed to empower dispatchers and responders with tools to make quicker, more accurate decisions, ultimately improving outcomes in emergency situations.

Features

1. Interactive Dispatcher Dashboard

Simulate realistic emergency call scenarios using AI models.
Get real-time feedback on responses.
Graded performance analytics based on scenario-specific expectations.

2. Responder Tracking

Locate and notify nearby first responders in real-time.
Use geolocation to connect responders with emergencies.
Monitor response progress and provide updates to dispatchers.

3. Persistent Chat and Feedback

Maintain a record of user responses for continuous improvement.
Provide comments and hints on missed steps or incorrect actions.

4. AI-Driven Scenarios

Use LangChain and OpenAI models to generate emergency scenarios dynamically.
Ensure scenarios align with best practices in emergency response.

Technology Stack

Frontend

Streamlit: Interactive UI for dispatcher training and responder tracking.

Backend

MongoDB: Stores persistent chat logs, user performance, and grading information.
LangChain: Manages the logic for conversational AI and scenario generation.
- Imports used:
- from langchain.prompts import PromptTemplate
- from langchain.chains import ConversationChain
- from langchain.memory import ConversationBufferMemory
OpenAI GPT-4: Generates realistic emergency scenarios and AI responses.

Additional Tools

pymongo: For MongoDB integration.
dotenv: Manages environment variables securely.
st-js-cookie-manager: Handles cookies in Streamlit for user session management.

Setup Instructions

Clone the Repository:

git clone https://github.com/your-repo/watchover.git
cd watchover

Set Up a Virtual Environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

Install Dependencies:
```
pip install -r requirements.txt
```

Configure Environment Variables: Create a .env.local file and add:

OPENAI_API_KEY=your_openai_api_key
MONGO_URI=your_mongo_connection_string

Run the Application:
```
streamlit run app.py
```

Future Enhancements

Multi-language Support: Train dispatchers in multiple languages.
Mobile App Integration: Extend real-time responder tracking to mobile devices.
Advanced Analytics: Provide deeper insights into dispatcher performance trends.