ChatGPT’s ability to understand whiteboards is absolutely bonkers.
1. **Main Objective**:
- "HOW TO AUTOMATE OVER 50% OF DOCTORS AND NURSES TIME AWAY FROM PAPERWORK AND CRAPPY SOFTWARE"
2. **Clinical Setting**:
- **ER Room**: Indicates an Emergency Room setting.
- There's a depiction of a doctor interacting with patients, with a note "Whisper -> CoreML" suggesting the use of the "Whisper" voice recognition tool using Apple's "CoreML" for real-time transcription.
3. **Documentation Tools**:
- "ResDoc" & "NOTE IDMD"
- An illustration of a device, perhaps a tablet or computer, with "TEPIC" and "EHR/EMR Preview".
4. **Data Processing & Transfer**:
- "techtop -> oodlsey bad"
- "stream" with a note of "2min 2000 tokens -> .7"
- FHIR API with a series of actions:
- "check patient exists"
- "submit report to eIRX"
- "get past reports (ideally)"
5. **Cloud & Security**:
- "SOC 2 AZURE"
- Notes around it include:
- "Azure Functions fire it"
- "Data QA"
- "api opt -> 35"
- "prompteng 2 for automatic query runtime"
6. **Middleware**:
- References to "Middleware" connecting different elements.
- Notes include "Studentson feeling chill", suggesting perhaps a specific middleware tool or a comment about the ease of its use.
7. **Data Structuring & Presentation**:
- "GraphQL Schema" with a note "Rz User -> main copy"
- "FRONTEND" which includes:
- "GraphQL Schema"
- "Summarization iOS App"
- "Dashboard app, etc."
- An additional note: "COPYPASTA Summarize Return Provider SERVICE".
8. **Data Storage**:
- Circles labelled "DD Postgress" and "Cosmos" suggesting database storage solutions.
9. **Miscellaneous**:
- Various arrows and lines connect different components, indicating data flow or integration points.
- There's a mention of "Summarizer Service (Function)" connected to middleware.
- Some obscured or partially visible words, such as "on..." or "usin..."
The whiteboard outlines a comprehensive system design focusing on automating documentation for healthcare professionals in an ER setting. It looks at integrating voice recognition with real-time data processing, adheres to healthcare data standards (like FHIR), and leverages cloud platforms and databases for scalability and data integrity.
I'd💞to thank
@DanielSnider from
@VectorInst for helping draft this open architecture for Canadian hospitals to deploy AI in a safe, scalable & auditable manner.
It enables both public & private owned code to interoperably solve the whole problem.
CC3.0 license💫Feedback welcome