Cognitive Disorder Detection in Adolescents using Mental Health Data

Cognitive disorders among adolescents pose significant challenges to their overall well-being and prospects. Timely detection and intervention are crucial to mitigating the potential long-term impacts of these disorders. This research aims to develop a framework that can detect multiple cognitive disorders in adolescents by first collecting subjects’ responses to various survey questions, which are derived from standard questionnaires used to assess the brain’s functioning, then utilizing the Conditional Tabular Generative Adversarial Network (CTGAN) for treating the imbalance in the dataset, and finally using the TabNet deep neural network for classifying disorders. The proposed framework outperforms traditional machine learning (ML) models and is a promising tool for the online detection of cognitive disorders in adolescents.