Skip to content

Repository files navigation

Ocunetra

Ocunetra is a mobile iOS app that uses AI and machine learning to help users screen for common eye conditions using only their smartphone

What the App Does

  • Screens for Diabetic Retinopathy using a model trained on 6,000+ images
  • Screens for Conjunctivitis using a model trained on 2,500+ images
  • Generates SHAP heatmaps to show what areas influenced the AI's prediction
  • Includes common eye tests: Amsler Grid, Ishihara color blindness, contrast sensitivity, and peripheral vision
  • Lets users find nearby specialists with a built‑in map feature
  • Provides eye‑health tips, exercises, and safety guidance

Languages/technology Used

  • Languages: Python, Swift
  • Platforms: iOS, CoreML, MacOS (for development)
  • Tools: SHAP explainability, as well as tips about eye health

Why We Built It

We were inspired by personal family experiences with preventable blindness and the lack of access to eye care in rural and underserved areas. With most people owning smartphones, we wanted to create an affordable and accessible way for anyone to check their eye health.

Challenges

  • Training and optimizing the ML models
  • Bridging SHAP outputs from Python into CoreML
  • Designing a simple, clear UI that anyone can use

Future Improvements

  • Multilingual and text‑to‑speech support
  • Secure clinical data sharing and remote monitoring features

Demo Video

https://youtu.be/izvR91hsc-I

Source Code

https://github.com/RonitBStudent/Final-Submission-CAC

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages