AI-Based Screening for Depression and Social Anxiety Through Eye Tracking: An Exploratory Study

StatusVoR
Dublin Core metadata
Alternative title
Authors
Chlasta, Karol
Wisiecka, Katarzyna
Krejtz, Krzysztof
Krejtz, Izabela
Monograph
Monograph (alternative title)
Date
2024-12
Publisher
Journal title
International Journal of Marketing, Communication and New Media
Issue
15
Volume
Pages
Pages
75-91
DOI
ISSN
2182-9306
ISSN of series
Weblink
http://u3isjournal.isvouga.pt/index.php/ijmcnm/article/view/925
Access date
2024-12
Abstract PL
Abstract EN
Well-being is a dynamic construct that evolves over time and fluctuates within individuals, posing challenges in its quantification. Reduced well-being is often associated with depression or anxiety disorders, characterised by biases in visual attention towards specific stimuli, such as human faces. This paper introduces a novel approach to AI-aided screening of these affective disorders by analysing scan paths of visual attention using convolutional neural networks (CNNs). Data were collected during two studies assessing (1) attention tendencies among individuals diagnosed with major depression and (2) social anxiety. These data were applied to residual CNNs through images generated from eye-gaze patterns. The experimental results, obtained using ResNet architectures, demonstrated a promising average accuracy of 48% for a three-class system and 62% for a two-class system. Based on these exploratory findings, we propose that this method could be utilised in rapid, ecological, and effective mental health screening systems to quantify well-being through eye-tracking.
Abstract other
Keywords PL
Keywords EN
Eye-tracking
Artificial intelligence
Convolutional neural networks
Depression
Social anxiety
Well-being
Keywords other
Sustainable Development Goals
SDG 5. Gender Equality ,
SDG 9. Industry, innovation and infrastructure ,
SDG 3. Good Health and Well-being
Exhibition title
Place of exhibition (institution)
Exhibition curator
URI
https://share.swps.edu.pl/handle/swps/1966
Type
License type
cc-by-nc
Except as otherwise noted, this item is licensed under the Attribution-NonCommercial licence | Permitted use of copyrighted works
Funder
Time range from
Time range to
Contact person name
Related publication
Related publication
Grant/project name
Files
AI-Based Screening for Depression and Social Anxiety Through Eye Tracking:
AI-Based Screening for Depression and Social Anxiety Through Eye Tracking:(579.34 KB)
Download
Views
Views5
Acquisition Date16.12.2025
Downloads
Downloads2
Acquisition Date16.12.2025
Altmetrics©
Dimensions
Google Scholar
Google Scholar