Integrative Machine Learning Approach for the Early Detection of Motor Dysgraphia via Handwriting and EEG Signal Analysis

A specific cognitive disability called motor dysgraphia impairs a student’s ability to write effectively, frequently resulting in significant scholastic difficulties. Due to the lack of standardized diagnostic techniques, this disorder is commonly disregarded despite its significance. For educators and medical professionals to properly support impacted pupils, early detection is essential for prompt action. This work uses machine learning and power spectral density (PSD) analysis to detect motor dysgraphia through a dual-modality technique that combines handwriting analysis and electroencephalogram (EEG) data processing. Kinematic attributes such as pen pressure, directional movement in x-and y-coordinates, pen height, azimuth, and on-surface versus in-air movement are used to gather and evaluate handwriting data. The underlying neurological variables are not explored by handwriting analysis alone, even though it offers initial insights. EEG signals from children who were identified as normal and dysgraphic by handwriting analysis are further analyzed to overcome this constraint. EEG recordings are subjected to Welch’s power spectral density (PSD) analysis to confirm previous hypotheses and identify patterns of brain activity linked to motor dysfunction. A greater comprehension of the connection between motor control, cognitive functions, and cerebral activity is made possible by the use of artificial intelligence models. To boost the early detection of motor dysgraphia and aid in the creation of focused therapy interventions, our suggested framework provides a highly accurate, AI-powered diagnostic tool. The findings show that an automated, non-invasive, and effective diagnostic method is feasible and can greatly enhance the educational experiences of kids with learning difficulties.