Research Projects

Dyslexia is a neurobiological, language-based learning disability affecting cognitive processes related to reading, spelling, and writing, with an estimated prevalence of 15% in India and worldwide. Dyslexic individuals often face difficulties in effectively communicating with others, such as establishing connections between words, sounds, and letters. Common symptoms include poor spelling abilities, slow reading speed, poor writing skills, inability to decode words mentally, difficulties pronouncing words while reading aloud, and poor reading comprehension. Despite global recognition, existing detection and intervention tools are predominantly English-centric, limiting their applicability in multilingual contexts like India. To address these challenges, technology-driven detection and intervention methods are essential. In this regard, eye-tracking technology, known for its linguistic neutrality, offers promise for developing targeted interventions to improve reading performance in dyslexic individuals across various linguistic environments. Furthermore, eye-tracking-based on-screen systems allow users to type with their eyes, and this technology could be leveraged as a dyslexia screening or detection tool by analyzing reading, spelling, and writing patterns. This project aims to develop and evaluate a multimodal gaze-controlled learning system integrating eye-tracking, surface electromyography, touch switches, and machine learning to support dyslexic children. The system will feature an on-screen virtual keyboard in Hindi, Gujarati, and English to identify dyslexia-related challenges and deliver tailored interventions. By leveraging adaptive tools and repeated exposure, it seeks to detect early warning signs, reduce symptoms, and improve learning outcomes. The research hypothesizes that combining eye-tracking measures with machine learning can effectively detect dyslexia-related oculomotor deficits and improve reading fluency through personalized, language-specific interventions. The project involves developing virtual keyboard interfaces, collecting and analyzing multilingual sensor data, developing robust machine learning models, and designing adaptive learning modules tailored to the cognitive demands of regional languages. The successful implementation of the iGazeBuddy project could enhance dyslexia understanding in multilingual contexts by providing scalable, language-agnostic diagnostic and intervention tools. It aims to improve educational support for dyslexic children in India and beyond, aligning with NEP 2020’s inclusive education goals. By enabling early detection and personalized interventions, iGazeBuddy can improve learning outcomes, especially in under-resourced regions, while supporting India's vision of enhancing educational accessibility, quality, and advancing assistive learning technologies.