Large Language Models in Mental Health: A Systematic Review of Applications, Innovations, and Ethical Challenges
06:00 · August 20, 2026 · arXiv cs.AI RSS
We present a review on the applications of large language models (LLMs) in health, e.g., social media analysis, clinical conversational agents, therapy support tools, prompt engineering, multimodal learning, and ethical considerations. We integrate findings from interdisciplinary studies utilizing diverse data sources such as social media posts, electronic medical records, and multimodal inputs to enable early detection of depression, suicide risk assessment, personalized therapy support, and psychoeducational content generation. Our review highlights advancements in LLM models and annotation strategies that enhance interpretability and clinical relevance, while we also emphasize the critical role of prompt engineering for domain adaptation. We also discuss emerging multimodal fusion techniques integrating text, speech, and sensor data for improved mental health diagnosis and monitoring. Finally, we address ongoing ethical, sociotechnical, and regulatory challenges, and advocate frameworks to ensure safe, equitable, and accountable deployment of LLMs in real-world mental health care.
Summary
A systematic review examines how large language models are being applied across mental health contexts, drawing on interdisciplinary studies that combine social media posts, electronic medical records, and other multimodal inputs. The work covers clinical conversational agents, therapy support tools, and automated analysis for early detection of depression and suicide risk, alongside generation of personalized therapy responses and psychoeducational material.
The review underscores progress in prompt engineering techniques that adapt general-purpose models to clinical language and tasks, as well as improved annotation strategies that increase interpretability and alignment with clinical standards. It also surveys emerging multimodal fusion methods that combine textual data with speech signals and sensor streams to support more continuous diagnosis and monitoring.
At the same time, the authors stress persistent ethical, sociotechnical, and regulatory obstacles to safe deployment. They call for structured frameworks that promote accountability, equity, and appropriate oversight before these systems move into routine mental health care.
Why it matters
The review directly aligns with the Dutch AI market's strong emphasis on ethical, transparent AI and its robust HealthTech sector. It provides researchers with a comprehensive overview of state-of-the-art multimodal techniques and regulatory frameworks necessary for deploying AI in sensitive domains like mental health under EU standards.
