Artificial intelligence robots for mental health applications: a scoping review

Area of Science:

• Digital Health
• Artificial Intelligence in Healthcare
• Mental Health Technology

Background:

• Growing demand for mental health services exceeds professional capacity.
• Artificial intelligence (AI) robots offer a promising solution for digital mental health interventions.
• AI robots present unique application potential in addressing mental health needs.

Purpose of the Study:

• Systematically analyze AI-robot characteristics and technical pathways in mental health.
• Guide developers in selecting appropriate AI technologies for mental health applications.
• Provide a basis for mental health providers and researchers to choose suitable AI robots.

Main Methods:

• Conducted a scoping review following JBI methodology and PRISMA-ScR guidelines.
• Searched major databases (PubMed, Embase, PsycINFO, Scopus, IEEE Xplore, ACM, Cochrane) for relevant literature.
• Included research on AI-robot applications in mental health published between January 1, 2017, and May 5, 2025.
• Utilized narrative synthesis to integrate evidence from 34 selected studies.

Main Results:

• AI-robots were applied in diagnosis (19 studies), monitoring (5), treatment/intervention (27), risk prediction (8), and counseling (13).
• Key research areas include anxiety, depressive, autism spectrum, and dementia disorders.
• Machine learning algorithms enhance diagnostic efficacy; Natural Language Processing (NLP) and generative AI improve personalized interventions.
• AI models generally perform well, with accuracy as the most reported metric.

Conclusions:

• AI-robots show promise in mental health services but face challenges in clinical efficacy, technical effectiveness, privacy, and ethics.
• Future research should optimize algorithms and conduct rigorous randomized clinical trials.
• Improving privacy and security mechanisms is crucial for safe, reliable, and sustainable AI applications in mental health.