NeuroRead – AI-Powered Early Dyslexia Detection

Inspiration

Dyslexia affects 3.5 crore children in India, yet early detection remains a challenge. Many children go undiagnosed for years, impacting their learning and confidence. Inspired by the need for an accessible, AI-driven solution, we created NeuroRead to help teachers and parents identify dyslexia early and provide personalised support.

What It Does

NeuroRead analyzes handwritten text to detect dyslexic patterns in children aged 6-12. Teachers and parents can upload a handwriting sample, and our AI model examines text structures, providing:

Early dyslexia detection
Detailed AI-generated reports
Personalized learning exercises
Continuous monitoring to track progress

By enabling timely intervention, NeuroRead helps educators and parents create a more inclusive learning environment.

How We Built It

We combined AI, handwriting recognition, and personalized learning to develop NeuroRead. The core steps included:

Image Processing: Extracting handwriting samples for analysis.
Machine Learning: Using CNN models to detect dyslexic patterns.
AI Reports: Generating detailed reports with Gen AI insights.
User Interface: Designing an easy-to-use platform for teachers, parents, and psychologists.

Challenges We Ran Into

Data Availability: Limited access to high-quality dyslexic handwriting datasets.
Data Privacy: Handling sensitive student data securely and ethically.

Accomplishments That We're Proud Of

Developing a working prototype that successfully detects dyslexia patterns.
Built the platform using publicly available open-source datasets at no additional costs.
Datasets
[https://universe.roboflow.com/embedded-uukr8/children-with-dyslexia-handwrite-recognization/dataset/2]
[https://www.kaggle.com/datasets/drizasazanitaisa/dyslexia-handwriting-dataset?resource=download]
Tackle Trivial Technical issues such as data storage and security using open-source solutions hence building a low-cost platform.
Ensuring an intuitive and accessible interface for non-technical users.
Building an AI-powered solution that can positively impact millions of children.
We have developed a dual-approach business model that enables us to reach students through both teachers and parents.

What We Learned

The importance of early intervention in dyslexia detection.
How AI and education technology can merge to create real-world impact.
Challenges in AI fairness and bias when dealing with sensitive educational tools.

What's Next for NeuroRead

Enhancing model accuracy with more diverse datasets.
Model Accuracy:Enhancing model accuracy with ground-truth & diverse dataset for real-world testing.
Developing a mobile app for wider accessibility.
Integrating multilingual support to expand reach across regions.
Creating an ecosystem of psychologists and therapists to validate our report and to provide easy access to professional help for parents.

NeuroRead aims to revolutionize dyslexia detection and make learning accessible to all. 🚀📖