AWS Builder Center
LifeLink Grid: Building an AWS-Powered Emergency Medical Coordination Layer

LifeLink Grid: Building an AWS-Powered Emergency Medical Coordination Layer

What if finding the right hospital during an emergency didn't require calling multiple hospitals and ambulance services? We built LifeLink Grid as a serverless AWS coordination layer that connects emergency requests, hospital resources, ambulances, and dispatch.

🚑 What If the Right Hospital Could Find

You?

When someone needs urgent care, families may have to call hospitals to check beds, contact blood banks, search for oxygen, and call ambulance services separately. The problem isn't necessarily the absence of information — it's the lack of coordination between systems.
We built LifeLink Grid to address that gap.
Not the nearest hospital. The right hospital.



🗣️ Built for Everyone — Not Just for

People Who Type Fast

One of the things we cared about was accessibility.
For many people, especially in rural communities, typing a detailed emergency request on a smartphone may not be the easiest way to ask for help.
So we designed LifeLink with voice input and a multilingual experience in English, Hindi and Kannada.
A person can interact with the emergency flow using voice instead of having to type everything manually.
The idea is simple:
If technology is meant to help during an emergency, people should be able to communicate with it in the way they are most comfortable with.
This isn't just a UI feature.
For us, it is about making digital emergency coordination more accessible across different languages, literacy levels and locations.



💡 The Idea

LifeLink Grid is an emergency medical coordination layer connecting citizens, hospitals, resources, and ambulances.
A citizen submits an emergency request with:
  • 🛏️ Bed
  • 🩸 Blood
  • 🫁 Oxygen
  • 🚑 Ambulance
  • 📍 Location
  • 🚨 Urgency Instead of simply finding the nearest hospital, LifeLink evaluates resource availability together with location to identify a suitable facility. Consider three hospitals:

Hospital A
Hospital A
Hospital A
Distance
3 km
6 km
4 km
Bed
✅
✅
❌
Blood
❌
✅
✅
Oxygen
✅
✅
✅
  • If a patient needs a bed + blood + oxygen, the closest hospital isn't necessarily the most suitable one.
That's why LifeLink looks at the whole requirement, not just distance.
We don't ask:

“What's the nearest hospital?”

We ask:

“Which available facility can actually help?

LifeLink goes beyond hospital search.
Suppose the best-fit hospital has the required medical resources — but no ambulance is available.
Instead of stopping there, LifeLink can coordinate an available ambulance from another participating facility.

Coordinate the resources around the patient, not just the hospital around the patient.



☁️ Under the Hood: AWS Powers the

Grid

We built LifeLink Grid using a serverless AWS architecture.

  • ⚡ AWS Lambda
Runs our modular backend functions for facilities, resources, emergency requests, dispatch, tracking and more.
  • 🌐 Amazon API Gateway
Provides the REST API connecting our frontend to the backend.
  • 🗄️ Amazon DynamoDB
Stores facilities, resources, requests, matches, redistribution, empanelment and trips.
  • ⏱️ Amazon EventBridge
Runs scheduled background workflows such as expired-hold cleanup and redistribution scanning.
  • 📢 Amazon SNS
Handles event-driven notifications such as incoming-patient and ambulance-commitment events.
  • 🤖 Amazon Bedrock
We built a triage Lambda using Claude Haiku 4.5. During the event, account verification prevented live model invocation, so our system uses a safe fallback rather than displaying fabricated AI output.
  • 🛠️ AWS SAM + CloudFormation
Our infrastructure is defined as code, allowing us to validate, build and deploy the backend consistently.



🏥 We Don't Replace Hospital Software.

We Connect It.

One question from our feedback particularly influenced our architecture:
“Hospitals already have their own software. Why would they use another system?”
Our answer:

They don't have to replace it.

LifeLink is designed as a coordination and interoperability layer.
We implemented a Hospital HIS adapter through:
 POST /interop/sync 
Hospital-side information can be translated into the common resource representation used by LifeLink.
So the architecture becomes:




🩸 Don't Just Wait for an Emergency

We also wanted LifeLink to think beyond the moment a request arrives.
Our background redistribution scanner can flag situations such as surplus or near-expiry blood inventory, creating an opportunity to redistribute resources before another emergency request needs them.
That changes the mindset from:
“Someone asked. Now we search.”
to:
“We identified an opportunity before someone had to ask.”



🔐 Built to Coordinate — Not to Replace

Human Decisions

LifeLink can automate matching, coordination and notifications, but critical medical decisions still require appropriate human oversight.
The system is designed to assist people and connect resources, not replace doctors or emergency professionals.
Our demo uses seeded data and does not claim live government feeds or live hospital inventory.
We believe being clear about what a prototype actually does is just as important as making it impressive.



🚀 What's Next?

Our hackathon prototype uses seeded demonstration data and does not claim live government feeds.
The next step is deeper interoperability with real healthcare systems, including standards-based integrations, richer routing, broader resource coordination, and expanded AI assistance with appropriate human oversight.
Because in an emergency, finding a hospital isn't enough.

You need the right hospital, the right resources, and the right coordination — at the right time.

That's LifeLink Grid. 🚑


Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article