Building CareGraph AI: Autonomous Multilingual Voice & Knowledge Graph Healthcare Triage with FalkorDB

Description and use

CareGraph AI uses FalkorDB to instantly identify critical conditions from voice input, enabling rapid ICU bed allocation and specialist assignment across India.

Problem

Emergency medical triage in India is slowed by hallucinating chatbots and disconnected hospital resources, causing critical delays in life‑saving decisions.

Outcome

The system achieves sub‑second deterministic triage, replacing hallucinating LLMs with a 0.31 ms FalkorDB traversal, enabling instant ICU bed allocation and ambulance dispatch across India.

Workflow

• Patient speaks symptoms via phone
• Deepgram Nova‑2 and Murf Falcon TTS process audio (<250 ms)
• Session and detected symptoms written to FalkorDB
• 5‑hop Cypher traversal identifies high‑risk condition, allergies, available ICU bed, and specialist
• Twilio PSTN dispatches ambulance CAD alert
• Hospital bed status updated to unavailable
• Automated follow‑up call confirms admission

Evidence

Traditional LLM chatbots guess medical advice probabilistically. In an emergency, a hallucinated drug recommendation or delayed triage is catastrophic.

Source

We replaced hallucinating LLMs with a 0.31ms 5‑hop FalkorDB traversal to connect voice symptoms directly to verified ICU beds and doctors across India.

Source

CareGraph AI unites FalkorDB, LiveKit WebRTC, Murf Falcon TTS, and Twilio PSTN into a closed‑loop emergency triage system

Source