AI Medical Triage System
Architected an AI-powered triage system for a Ghana hospital—streamlining patient-to-diagnosis flow using kiosks, AI diagnosis tools, and medical transcription.
The Challenge
Hospitals in Ghana faced long wait times and inefficiencies in patient intake and triage—patients spent hours before seeing a doctor.
Patient data was often poorly captured or lost, making follow-ups and diagnosis inefficient.
The hospital wanted to use AI to assist with triage and diagnosis—but needed a system that would work in a real-world setting with limited infrastructure.
The project required integrating multiple AI tools (Corti for diagnosis, Mediktor for transcription) into a unified workflow.
My Role
Technical architect and lead developer—responsible for designing the end-to-end solution, selecting and integrating AI tools, building the patient data platform, and creating the proposition that would secure funding.
What I Did
- Designed the full system architecture—patient kiosks for intake → centralized patient database → AI diagnosis assistance (Corti AI) → medical transcription (Mediktor) → doctor review → diagnosis.
- Built the patient data platform using Laravel and Vue.js to maintain a centralized, accessible database of patient records.
- Integrated Corti AI to assist with diagnosis—providing doctors with AI-powered insights to streamline the triage process.
- Integrated Mediktor for medical transcription, ensuring diagnosis details were accurately logged and stored.
- Designed kiosk workflows that captured patient details quickly and accurately, reducing manual intake errors.
- Created the technical proposition and cost estimates that would be presented to funders and hospital administrators.
The Result
- Delivered a complete technical proposition for an AI-powered triage system at a Ghana hospital.
- Designed a solution that would reduce patient wait times and improve diagnosis accuracy.
- Integrated cutting-edge AI tools (Corti, Mediktor) into a practical healthcare workflow.
- Created a centralized patient database that would enable better follow-ups and continuity of care.
- Built a system that could work in a real hospital environment with limited infrastructure.
- Project is currently awaiting funding—ready to implement as soon as resources are secured.
AI in healthcare isn't about replacing doctors—it's about making doctors better, faster, and more accurate. This project was about bringing that vision to a Ghana hospital where patients waited too long and data was too fragmented. The proposition is ready, the architecture is built, and the funding is the only thing standing between concept and reality.
The Technology
My Takeaway
“AI in healthcare isn't about the technology—it's about the workflow. You can have the best AI tools in the world, but if they don't fit into how doctors actually work, they're useless. Healthcare tech isn't about being impressive—it's about being useful.”