11-Year-Old Kerala Girl Develops AI App That Detects Eye Diseases

An 11-year-old girl from Kerala named Leena Rafeeq has invented an application based on artificial intelligence (AI) that detects eye diseases and other eye conditions through a unique scanning technique using an iPhone. Rafeeq named this application "Ogler EyeScan." She was 10 years old when she developed this application.

Rafeeq explained how the application works on LinkedIn with a video. She said her application can analyse light and colour intensity and look-up points to locate the eyes within the range of the frame with the help of advanced computer vision and machine learning. The Ogler Eye Scan can determine whether the eyes are correctly positioned inside the scanner frame and can identify light burst issues as well.

She added that the Ogler Eye Scan can also identify any light burst issues and determine if the eyes are positioned exactly inside the scanner frame.

Kerala Girl Develops AI App

After the eyes are scanned, potential eye diseases or conditions like arcus, melanoma, pterygium, and cataract can be diagnosed using trained models. Rafeeq said that the app was developed using Swift UI without any third-party libraries or packages. It had taken six months of research and development for Rafeeq to create this app.

Rafeeq said that she learned more about various eye conditions, computer vision, machine learning models, algorithms, and advanced levels of Apple iOS development, including sensor data, AR, Core ML, Create ML, etc.

This Ogler Eye Scan app is only supported on iPhone 10 and above with iOS 16+. Currently, the app is under review by the App Store, and Rafeeq hopes that it’ll get listed soon.

Many netizens were amazed by such an impressive innovation by the 11-year-old and congratulated her. "Great work! Good example of how we could prevent equity in health using AI," wrote one user. "You deserve huge congratulations! Great job at the age of 10!" commented another user. A third user wrote, "You’ve achieved so much with Ogler Eye Scan. My best wishes!"

In response to one user, Rafeeq said that the model is 70 percent accurate. She explained that the distance required for capturing scans through phone devices is caused by the presence of glare and bursts from lights. Rafeeq added that they had implemented metrics and detection for light-related issues. If there is a light-related issue, the user is prompted to take a re-scan.

Rafeeq also mentioned that she is currently concentrating on training more sophisticated models. Updates will be released once Ogler Eye Scan is accepted by the App Store.

Suggested Reading: 10 Things To Know About Gitanjali Rao, Teen Inventor And Time Magazine’s Kid Of The Year