Noa — Healthcare That Remembers

Inspiration

Healthcare conversations contain some of the most valuable information about a patient's condition, but much of that knowledge is lost after every consultation.

A patient may explain their symptoms, previous experiences, concerns, and medical journey to one doctor, only to repeat the same story when they visit another provider. Doctors often spend valuable consultation time collecting information and documenting notes instead of focusing on care.

We were inspired by a simple question:

What if healthcare could remember every conversation?

Noa was created to build a continuous memory layer for healthcare — preserving patient stories and transforming them into actionable clinical context.

---

What We Built

Noa is a voice-first medical intelligence platform with two core experiences:

1. Patient Voice Intake

Before a consultation, patients can open a simple web interface and speak naturally about their symptoms.

Instead of filling forms, they have an AI-guided conversation where Noa asks adaptive follow-up questions, understands context, and captures their medical story.

2. Doctor Consultation Intelligence

During consultations, doctors can record conversations with patients. Noa captures the discussion, creates a transcript, and converts the conversation into structured clinical documentation.

The system generates:

• SOAP clinical notes
• Patient summaries
• Triage scores
• Red flag indicators
• Doctor briefings

This gives clinicians the context they need before making decisions.

---

How We Built It

Noa was built using AWS services:

• Amazon Bedrock Nova 2 Sonic for real-time voice conversations and contextual interaction
• Amazon Bedrock Nova 2 Lite for clinical reasoning and structured summaries
• Amazon DynamoDB for storing patient profiles, consultation sessions, transcripts, and medical memory records
• Amazon Cognito for authentication and user management
• Amazon S3 for storing session recordings and generated files

The application was built with:

• Next.js
• TypeScript
• Tailwind CSS
• WebSocket-based realtime communication

The architecture allows every consultation to become part of a patient's longitudinal health memory.

---

Challenges

One of the biggest challenges was designing a system that could handle natural human conversations while still producing structured clinical information.

Medical conversations are unpredictable. Patients describe symptoms differently, provide incomplete information, and often explain problems in their own words.

We focused on building a system that balances conversational flexibility with structured outputs that are useful for healthcare professionals.

Another challenge was designing a workflow that serves both patients and doctors — making the patient experience simple while providing doctors with meaningful clinical insights.

---

What We Learned

Building Noa showed us that the future of healthcare AI is not only about generating answers — it is about preserving context.

The most valuable information is often hidden inside conversations. By combining voice AI, clinical reasoning, and scalable cloud infrastructure, we can create healthcare systems that remember, understand, and support better decisions.

Noa is a step toward healthcare where every patient story matters and no important context is lost.

Built With

• amazon-dynamodb
• bedrock
• nextjs