NAGOMI: Health-Assistant Voice Agent

Nagomi is the voice-first telehealth platform that bridges patients and multi-clinic networks—automating admin, enhancing care coordination, and giving clinicians back their time.”

Comment

Nagomi is the voice-first telehealth platform that bridges patients and multi-clinic networks—automating admin, enhancing care coordination, and giving clinicians back their time.

In Japanese, Nagomi (なごみ / 和み) carries a rich and serene meaning. It is derived from the verb nagomu (和む), which means to calm down, to be comforted, or to become peaceful. At its core, Nagomi represents a state of inner peace, harmony, and tranquility, both within oneself and in relation to the world.

Inspiration

  • Clinician burnout from endless EHR clicks and after-hours charting
  • Fragmented patient journeys across dispersed clinics
  • Accessibility barriers for seniors, rural populations, and non-English speakers
  • The rise of voice-first experiences in consumer tech, untapped in healthcare

What it does

  • Patient Side:
  • Voice-driven symptom triage & intake
  • Appointment booking, reminders, rescheduling
  • Medication and follow-up reminders
  • Multilingual, offline-capable health education
  • Clinician Side:
  • Real-time voice-to-text charting with FHIR/EHR integration
  • AI-powered CPT/ICD coding suggestions
  • Cross-clinic care plans & voice-activated referrals
  • Secure voice messaging and task management

How we built it

  • Frontend: React Native mobile app

  • Backend : ElevenLabs AI + Supabase

  • AI Stack:

  • Speech-to-Text: On-device Whisper models + cloud fallback

  • NLP & Intent: Custom-tuned transformer models trained on medical dialogs

  • Text-to-Speech: Neural, clinic-branded voices per locale

Challenges we ran into

  • User Trust: Ensuring patients and clinicians feel confident in AI-driven suggestion

Accomplishments that we're proud of

  • Multi-Lingual Roll-out: Launched in English, Spanish, Vietnamese with region-specific health dialogs
  • Clinical Validation: Achieved 92% symptom-triage accuracy against nurse-administered surveys

What we learned

Clinical Validation

What's next for NAGOMI: Health-Assistant Voice Agent

  • Payer Partnerships: Launch value-based care pilots to demonstrate reduced ER visits and lower costs
  • Predictive Analytics Module: Early warning alerts for readmission risk, based on voice-captured vitals and symptoms

Built With

Share this project:

Updates