Year: 2025
Type: dissertation
Abstract: Mental health issues are some of the biggest challenges that impacts over a billion people and healthcare systems everywhere. The COVID-19 pandemic made things worse, highlighting serious gaps in access to timely and effective care. This study sets out to explore how artificial intelligence (AI) tools improves mental health support and treatment. The aim was to pinpoint which AI applications provide the most practical, ethical, and scalable solutions to existing mental healthcare services. To conduct the study, a structured comparative analysis was performed on various AI technologies in the mental health space, including chatbots, emotion recognition systems, and multimodal sensing tools. A scoring framework was developed to evaluate these tools based on key factors like functionality, user satisfaction, clinical validity, implementation costs, transparency, and ethical compliance. The study focused on publicly available, non-clinical AI applications, steering clear of experimental models that haven't been deployed in real-world settings or tools meant solely for institutional use. The findings revealed that rule-based chatbots, such as Woebot and Tess, strike the best balance between effectiveness, accessibility, and ethical considerations. On the other hand, more advanced systems like multimodal AI models showed impressive diagnostic accuracy but faced challenges related to cost, data privacy, and infrastructure needs. This research sheds light on how AI can be thoughtfully integrated into mental health care. It provides a solid framework for stakeholders to assess new technologies and make informed choices about future implementations.
Source: Theseus (Ammattikorkeakoulujen)
Author: Kaur, Maninder
Institution: Helsinki Metropolia University of Applied Sciences
Language: English
Subfield: Applied Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7035595927
