About Lexora
Lexora is an eye-tracking-based dyslexia screening platform. It combines gaze analysis with machine learning to provide accessible, non-invasive reading difficulty detection.
Our Mission
Dyslexia affects approximately 10-15% of the population, yet many cases go undiagnosed for years. Traditional screening methods can be expensive, time-consuming, and inaccessible.
Lexora aims to change this by providing a fast, research-backed screening tool that anyone can use — whether they have access to professional eye-tracking hardware or just a standard webcam. Our platform analyses how eyes move during reading to identify patterns commonly associated with dyslexia.
How It Works
Eye Tracking
Supports Tobii hardware for research-grade precision and webcam-based tracking via MediaPipe for accessibility.
Gaze Analysis
Extracts fixation duration, saccade patterns, regression frequency, and other features linked to reading difficulty.
ML Classification
A trained machine learning model evaluates the extracted gaze features and produces a risk-level assessment.
Privacy First
All processing happens locally in your browser. No video is stored. Gaze coordinates are abstract and anonymized.
Important Disclaimer
Lexora is a screening tool developed for research purposes. It does not provide a clinical diagnosis of dyslexia. Results should be interpreted by qualified educational psychologists or healthcare professionals. This tool is not a substitute for professional assessment.
Research Context
This project is developed as part of a research initiative at the Faculty of Computing and Data Science. It draws on established eye-tracking research in reading and dyslexia detection, including work by Rayner (2009) and Holmqvist & Andersson (2017).