Voice assisted medical triage
The application combines a voice interview with patient information to prepare a clinician summary, patient slip and queue entry.
Amazon · Singapore · 25 March 2026
From voice intake to a clinician handoff
The interview combines with selected conditions, allergies and medications. These inputs feed a JSON ticket containing a summary, symptoms, assessment, recommended actions, clinician notes and a P1 to P4 acuity category.
Distinct records and queue state
Patient, conversation, ticket and queue records connect intake to waiting, in consultation and completed states. A printable slip and queue view carry the handoff beyond the voice conversation.
Hosted connections
A React and TypeScript client uses separate Bedrock paths for voice and ticket generation. Amplify Data and DynamoDB connect the records using public API key authorisation.
Scope
This educational application is not clinically validated or ready for real patient data. Clinical validation and clinician access controls require further work.
Team project
Richard Lee built dtax with Anni, Darryl and Zhi Kai at an Amazon hackathon and shared the work at the Amazon Web Services (AWS) User Group. The presentation discussed Cognito and Lambda as architecture options.