Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
AI in psychiatry: Perspectives of patients from Southeast Europe on ChatGPT
0
Zitationen
6
Autoren
2025
Jahr
Abstract
The integration of Artificial Intelligence (AI) into healthcare presents new possibilities and challenges. Large Language Models have shown potential in various psychiatric applications. However, the perspectives of patients with mental disorders on the use of such technologies remain underexplored. The present study aimed to evaluate the perceptions of patients with diagnosable mental disorders regarding the advantages and drawbacks of using ChatGPT for acquiring information about their conditions and medications. The data were collected at major psychiatric centres in Croatia and Bosnia and Herzegovina throughout October 2023. The sample consisted of 89 outpatients. The procedure involved inviting outpatients to participate in a questionnaire-based study that assessed their internet access, prior use of ChatGPT, and, after using ChatGPT to inquire about their mental health conditions and medications, their experiences interacting with ChatGPT. Data were analyzed using descriptive statistics, chi-square tests, t-tests, and logistic regression. The study found that 47.2% of the participants had used ChatGPT before. The main advantages noted were ChatGPT’s availability and immediate response capability. However, significant drawbacks included the lack of personal contact and the generality of the responses. Participants expressed concerns about the quality and specificity of information regarding their medical conditions. While ChatGPT offers notable advantages such as accessibility and promptness, the lack of emotional engagement and the sometimes vague nature of its responses limit its effectiveness from the patients' perspective. These findings suggest a need for enhancements in AI technologies to better address the unique needs and preferences of psychiatric patients.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.231 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.084 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.444 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.776 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.423 Zit.