🧠 Inspiration

We noticed a widespread issue: people often avoid going to the doctor when they suspect they have an STI—mainly due to embarrassment, stigma, or fear of showing their genitals. This leads to delayed diagnoses and untreated conditions. We wanted to build a solution that makes sexual health care more accessible, private, and stigma-free—especially for those who hesitate to seek in-person care.

🩺 What it does

Gonobee is the world’s first multimodal STI self-diagnosis app.
Users can perform a rapid, AI-powered check for six common STIs by using their camera and voice—with no need to log in or see a doctor.
The app provides the most likely condition based on input and offers guidance if medical care is recommended. All data is auto-deleted immediately after diagnosis.

🛠 How we built it

We combined image recognition and natural language processing via a multimodal AI architecture.

  • ElevenLabs was used for voice input processing.
  • ICD-11 codes and structure decision logic based on both symptoms (audio/text)

🧱 Challenges we ran into

  • medical accuracy The mock app showed the same answer regardless of the input. Training the real medical data again and again made this application finally work.
  • UX simplicity There were unnecessary buttons a lot first. We overcame this by providing design guidelines in the first prompt.
  • Ensuring privacy was the most critical part. Not only deleting, but we also considered how we can give users reassurance through any UI.

🏆 Accomplishments that we're proud of

  • Built the world’s first multimodal STI diagnostic tool.
  • Designed a judgment-free experience that feels safe and private.
  • Created a working prototype with real-time image & voice analysis.

🚀 What's next for Sexual Disease Diagnosis App Gonobee

We aim to:

  • Expand beyond STIs into internal medicine diagnosis using AI.
  • Partner with clinics for referral pathways and telemedicine integrations.
  • Improve multilingual and accessibility features.
  • Train on more diverse datasets to improve inclusivity and accuracy.

Built With

  • icd
Share this project:

Updates