Pulse Logo
Entry Page and Patient
Staff Interface
Doctor and Admin Interface

Pulse

Real-Time Healthcare Capacity and Intelligence Platform

Inspiration

Pulse was inspired by a real family emergency. In May, one of our family members had an accident and urgently needed ICU care. We traveled from one town to another searching for a public hospital that could admit the patient. There was no way to know which hospital had available beds, ICU capacity, or doctors on duty. We relied on phone calls, referrals, and guesswork while time was critical.

This experience revealed a major gap in the healthcare system. There is no platform where citizens can see real-time hospital availability, and no unified system for hospitals and local governments to monitor and manage healthcare capacity using historical data, existing records, and artificial intelligence.

What it does

Pulse is a real-time healthcare capacity, navigation, and intelligence platform designed for both citizens and healthcare administrators.

For Citizens

View nearby hospitals through an interactive GIS map
Check real-time bed, ICU, and doctor availability
Receive assistance through an AI health chatbot
Navigate to the most suitable hospital using optimized routing

For Hospitals and LGUs

Update and monitor hospital capacity in real time
View predictive analytics and congestion forecasts
Receive automated alerts when capacity thresholds are reached
Use a digital twin simulation to model emergency scenarios and patient flow

How we built it

Pulse was built as a cross-platform mobile and web system combining GIS, real-time databases, machine learning, and Gemini 3 as the intelligence layer.

Core Application

Flutter for the mobile app interface
Node.js for backend services and API handling
Firebase for authentication and real-time database updates
PostgreSQL for structured hospital and user data

GIS and Mapping

OpenStreetMap and Overpass API for hospital geolocation data
Routing and proximity calculations for hospital navigation

Machine Learning and Analytics

Python, Pandas, and Scikit-learn for analyzing historical hospital data
Predictive models for forecasting hospital congestion and resource demand
Analytics dashboards for administrators

Gemini 3 Integration (Core Intelligence Layer)

Gemini 3 plays a central role in transforming Pulse from a monitoring tool into an intelligent decision support system.

We used Gemini 3 in four major ways:

AI Health Chatbot Gemini understands user concerns written in natural language and connects this understanding with real-time hospital data to recommend appropriate hospitals.
Healthcare Data Interpretation Gemini processes live hospital capacity updates and converts raw numbers into meaningful summaries, alerts, and recommendations for the City Health Office.
Predictive Analytics Explanation While machine learning models generate forecasts, Gemini explains these predictions in clear, human-readable insights that administrators can act on.
Digital Twin Simulation Assistance In simulated emergency scenarios, Gemini interprets the results and summarizes possible outcomes and response strategies for decision-makers.

Gemini acts as the bridge between complex data, AI models, and human understanding.

Challenges we ran into

Accessing reliable and structured hospital data for mapping and capacity tracking
Designing a system useful for both citizens and healthcare administrators
Integrating real-time updates while maintaining a simple and intuitive interface
Explaining complex analytics and simulations in a way that non-technical users can understand
Ensuring scalability for city-wide use within hackathon constraints

Accomplishments that we're proud of

Developing a working prototype that demonstrates real-time hospital mapping
Successfully integrating Gemini 3 for both user assistance and administrative insights
Designing a digital twin concept for healthcare capacity simulation
Creating a solution based on a real healthcare problem experienced firsthand
Building a system designed for LGU integration and real-world adoption

What we learned

We learned how critical real-time healthcare data is during emergencies and how difficult it is to access without a centralized system. We gained experience in combining GIS, AI, analytics, and mobile development into a unified platform. We also learned how to design technology that supports both everyday users and institutional decision-makers.

What's next for Pulse

The next steps for Pulse include onboarding more hospitals, improving AI forecasting accuracy, expanding the digital twin simulation, and fully integrating the platform into the My Naga App. We also aim to include barangay health centers and make the system scalable for other cities and municipalities.