In-Depth Exploration of Advanced Machine Learning and Deep Learning Techniques for the Detection and Diagnosis of Dyslexia: A Comprehensive Review of Current Approaches, Challenge, and Future Direction

Myriads of neural regions that deal with linguistics in the brain getting disrupted will result in creating a specific learning disability known as dyslexia, that makes reading, writing, and spelling problematic. These people also face challenges with primary language acquisition. Primary dyslexia, or Developmental Dyslexia, Trauma Dyslexia, Visual Dyslexia, Auditory Dyslexia Secondary, and Dysgraphia make out for the six types of difficulties that especially stroke or war survivors tend to develop. Efforts to research and identify dyslexia have benefitted from the adoption of machine learning and deep learning technologies. They have further developed intricate advanced deep learning techniques which include the use of Convolution Neural Networks (CNN), and Long Short Term Memory (LSTM), Recurrent Neural Networks (RNN) networks for visual and sequential data processing that is linked to dyslexia detection. Mycosis aided hyperparasitism has also been studied with the aid of Support Vector Machines and Neural Networks. Metric saccadic eye movements, fixation duration, facial feature detection, and eye tracking all attribute towards determining dyslexia. This paper investigates several dimensions of difficulty in the detection of dyslexia, pinpointing the missing links and problems as well as areas of potential development. It also prefers for the application of other advanced techniques of deep neural networks plus other machine learning techniques.