Research Article | Open Access
Volume 2023 - 2 | Article ID 297 | https://dx.doi.org/10.51521/JMHPH.2023.e21.103
Academic Editor: John Bose
1Damilola Sherifat Shaba, 2Ridwan
Adebowale Yusuf, 3Taofeek Akinwumi Raheem, 4Sodiq Abiola
Omotosho
1Department of Public
Health, University of Illinois Springfield. Illinois, USA. ORCID: 0009-0008-7915-1122
2Department of Social
Science, Acap University College. Victoria, Australia. ORCID:
0009-0008-3329-0223
3Department of
Business Management, Eastern Gateway Community Colleg. Ohio, USA.
ORCID:
0009-0004-0593-0386
4Department of
Verterinary Science and Public Health, Fort Valley State University. Georgia,
USA.
Corresponding Author: Damilola Sherifat
Shaba, Department of Public Health, University of Illinois Springfield.
Illinois, USA. ORCID: 0009-0008-7915-1122.
Citation: Damilola Sherifat
Shaba, Ridwan Adebowale Yusuf, Taofeek Akinwumi Raheem, Sodiq Abiola Omotosho
(2023) AI and Mental Health Among Adolescents and Young Adults: Investigating
AI Models to Understand Trends, Risks, and Preventive Interventions in Schools
and Social Settings. J Ment Health Phys Health, 2(1);1-18.
Copyright: © 2023, Damilola
Sherifat Shaba, et al. This is an open-access article distributed under the
terms of the Creative Commons Attribution 4.0 International License, which
permits unrestricted use, distribution and reproduction in any medium, provided
the original author and source are credited.
Background: Adolescent and
young-adult mental health is a global public-health priority. Most mental
disorders begin before age 24, with roughly half
of lifetime cases starting by mid-adolescence and up to 74% by age 24 (Kessler
et al., 2005) [1]. Depression and anxiety are prevalent among college students,
yet an estimated 75% of those needing help do not access services due to stigma
and other barriers (Hunt & Eisenberg, 2010, as cited in Fitzpatrick et al.,
2017) [2]. Early interventions often last only ~2 years, and up to 80% of
youths may relapse after services end [3].
Objective: This study aims to
develop AI models to identify mental-health trends among adolescents/young
adults, assess risks (e.g. privacy, bias, stigma), and propose preventive
interventions in schools and social settings.
Methods: We outline a
multi-phase mixed-methods design. Participants (ages 12–24) will provide survey
data, social-media text, and school records. Natural language processing (NLP)
and machine-learning algorithms (e.g. classifiers) will detect sentiment and
risk patterns. Model performance will be evaluated by accuracy,
precision/recall, and fairness metrics.
Results: We anticipate
identifying key prevalence patterns (e.g. demographic differences), achieving
moderate predictive performance (expected F1 ~0.8), and uncovering risk factors
(e.g. bullying, social media use) correlating with distress.
Conclusions: AI can augment early
detection and support for youth mental health, but ethical safeguards privacy
protection, bias mitigation, and human oversight are crucial to ensure these
tools are safe and effective complements to traditional care.
Keywords: Adolescent Mental Health, Artificial intelligence, Machine learning, Preventive interventions.
.png)
.gif)
.jpg)
