Remedial learning assessment for dyslexic children in primary schools using deep learning

Dyslexia is a learning disability that hinders children from recognizing the letters and the words of a language. This results in poor reading, writing and linguistic skills in dyslexic children. In today’s revolutionary society, the increasing prevalence of dyslexia in schools has become a major concern for educationalists as it can lead to lifelong conditions of illiteracy and social exclusion. Dyslexia is difficult to diagnose in young children since its symptoms and indicators are not always clear and evident. Though there have been numerous diagnostic methods to screen for dyslexia, there is an acute need for proper educational intervention to facilitate the dyslexics to lead a dignified life. Multi-sensory learning technologies facilitate learning for dyslexic children. With the rapid development of Deep Learning techniques, suitable models can be used to provide learning evaluation and assessment for dyslexic students on the easy-to-handle edge devices. The mobile technologies can shift the monotonous classroom setting to more interactive sessions by providing a multisensory environment. The primary objective of this research work is to develop a mobile app for dyslexic students for remedial writing assessment of Tamil Letters using deep learning models. In this regard, deep learning models are proposed to recognize and evaluate dyslexic handwriting with appropriate assessment scores. Based on these scores, the learning of the students can be assessed.