Leveraging Artificial Immune Systems for Mental Health Re-search: Anomaly Detection in EEG Data

Bhandari, Shivani and Buckley, N and Secco, Emanuele Lindo (2025) Leveraging Artificial Immune Systems for Mental Health Re-search: Anomaly Detection in EEG Data. Artificial Intelligence and Applications. ISSN 2811-0854 (Accepted for Publication)

Text 4579 AIA.pdf - Accepted Version Restricted to Repository staff only Download (549kB)

Abstract

Mental Health is a physical, mental and social state affecting 970 million people in the world. Artificial Intelligence and deep learning techniques classifying ElectroEncephaloGraphy (EEG) data have emerged as a promising technology for the detection of mental health disorders. In this context, one underexplored area is the application of Artificial Immune Systems, which is a technique inspired by the human immune system that has been useful in many computational tasks, including anomaly detection. This paper aims to bridge the gap by leveraging Artificial Immune Systems for Mental Health through anomaly detection in EEG Data: a novel Negative Selection Clonal for Anomaly Detection (NSCAD) algorithm is presented and applied on a data set of 945 samples with individuals diagnosed with disorders and a control group of healthy participants. Efficacy of NSCAD on anomaly detection was assessed using precision, recall, F1-score, and accuracy metrics. Results are promising, with a precision of 0.92, a recall of 0.83, an F1-score of 0.88 and an accuracy of 0.78. A comparative analysis between the evaluation metrics and anomaly detection of NSCAD vs other methods is finally reported together with a critical analysis of the limitations.

Item Type:	Article
Faculty / Department:	Faculty of Human and Digital Sciences > School of Computer Science and the Environment
Depositing User:	Emanuele Secco
Date Deposited:	28 Apr 2025 11:02
Last Modified:	28 Apr 2025 11:02
URI:	https://hira.hope.ac.uk/id/eprint/4648

Actions (login required)

View Item