LUCID

Overview

I’m building LUCID with a University of Utah student team to help families recognize potential warning signs after a head injury through an ordinary phone call. Our functional prototype asks structured questions, remembers earlier conversations across interrupted calls, and calls back to check for changes. Callers can use a landline without an app or internet on their phone; the backend relies on computing and telecommunications infrastructure.

Role: AI architecture and engineering

Collaboration: Team: Walter Shewmake, David Ross, and Rodney Courtright. I built the software system; David developed the clinical decision logic.

Conversation and decision logic

The prototype design separates conversational AI from an explicit clinical rule set. AI interprets the caller's responses, while the rule set handles decision and escalation logic. My software work connects this flow with persistent conversational context, follow-up calls, and routing; David developed the clinical decision logic.

From prototype to local adaptation

We’re working with Glial Initiative on a Nigeria-specific implementation and sustainability proposal for Wema Hackaholics’ Social Impact track. I develop the software, while Glial contributes local research, community engagement, and stakeholder introductions. Together, we’re exploring language needs, telephone access, referral pathways, and an operating and funding model.

Building the organization

We’re working toward establishing a nonprofit and founding a mission:BRAIN chapter at the University of Utah. LUCID’s website now brings together a recorded prototype demo, an interactive sample conversation, and our presentation as we develop the project beyond the hackathon.

Prototype design

• 1Multilingual voice interaction and structured questions about potential warning signs
• 2Conversational context retained across interrupted calls
• 3Automated follow-up calls
• 4Escalation and call-routing features when available

Outcome & Status

Our three-person team won first place at the 2026 mission:BRAIN Global Neurosurgery Hackathon. We’re now developing the prototype and planning local adaptation. LUCID is not a clinically validated or deployed medical service; clinical review and evaluation planning are needed before pursuing a supervised pilot.