Smart Wearable Devices for Monitoring Cognitive and Behavioural Disorders in Students

Abstract

The rising rate of the cognitive and behavioural disorders amongst the students has necessitated the emergence of sophisticated monitoring systems that would help in detecting the disorders early and taking appropriate measures. The suggested architecture will be based on a multi-layered scheme where the data acquisition, communication, and data processing layers, and the application layer will be engaged. It does this by using data preprocessing and feature extraction algorithms to improve quality of data and machine learning (ML) and deep learning (DL) algorithms like Support Vector Machines (SVM), Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to classify and predict cognitive states. The results of the experiments indicate that deep learning models, particularly, the hybrid CNN-LSTM models are better in recognising stress, anxiety, and behavioural abnormalities compared to the traditional methods. Irrespective of the challenges in terms of data privacy, practicality of the appliances and generalization of the model, the proposed framework will provide a viable and scalable solution to track the mental condition of the students. The research can be used in the creation of intelligent healthcare and educational technologies, since it offers an in-depth, AI-based wearable monitoring system, which is student-specific. The provided plan can result in improved diagnostic outcomes, improved grades, and the health of students in general.