Image-Based Cognitive States Identification through Deep Learning
Shruti Rao, Siddarth Ambannavar, Ganesh Ram Sinha · 2024
Yoga and mindfulness practices offer a variety of benefits, promoting both physical and mental well-being. This research intersects with AI, psychology, neuroscience, and more within cognitive science, contributing to the study of the brain, mind, memory, and disorders. The study presents a novel approach utilizing deep learning and image classification techniques to distinguish practitioners from non-practitioners of yoga based on pre-processed data. Data pre-processing involves down sampling, filtering, and Independent Component Analysis (ICA) to enhance the extraction of relevant features. The pre-processed data is transformed into images using Gramian Angular Field, facilitating the utilization of Convolutional Neural Network (CNN) models. The proposed methodology leverages custom-built CNN models to classify individuals into their respective groups. The models are trained on the image representations of the pre-processed data, enabling the extraction of intricate patterns and features for accurate classification.