Researchers at the University of Illinois Urbana-Champaign have developed an innovative framework that uses generative artificial intelligence (AI) to advance personalized mental health treatment. The study, led by social work professor Cortney VanHook, demonstrates how AI can simulate real-world care journeys, helping clinicians and researchers better understand access barriers, improve cultural sensitivity, and enhance treatment outcomes.
Using the AI platform, the team created a detailed, simulated case for a fictional client, Marcus Johnson, a young Black man experiencing depressive symptoms. The AI analyzed Johnson’s personal and social context, including his support network, workplace stress, and cultural barriers to care.
Personalized treatment
The system then generated a personalized treatment plan using evidence-based frameworks such as Andersen’s Behavioral Model, the Five Components of Access, and Measurement-Based Care.
By applying these models, the researchers showed how AI can integrate patient data, theoretical models, and clinical reasoning to produce realistic treatment pathways. VanHook emphasized that the approach provides a safe, privacy-compliant environment for clinicians, students, and trainees to explore interventions without using real patient data.
“This work moves beyond theory,” said VanHook. “It’s a practical, evidence-based way to integrate AI into mental health education and practice, helping clinicians understand populations they may not often encounter and improving culturally competent care.”
AI’s potential in mental health access
The study also highlights AI’s potential to reduce inequities in mental health access. As all three authors, VanHook, Daniel Abusuampeh (University of Pittsburgh), and Jordan Pollard (University of Cincinnati), identify as Black men, they focused on ensuring the simulation accurately represented the systemic and cultural barriers Black men often face in the U.S. mental health system.
However, the researchers caution that AI is limited by the data it’s trained on, which may not fully reflect the complexity and emotional nuance of clinical encounters. “Generative AI can’t yet capture every factor influencing mental health care,” VanHook noted, “but it can help us better understand and design more equitable systems.”
The study arrives amid growing scrutiny of AI in health care. Under Illinois’ new Wellness and Oversight for Psychological Resources Act, AI tools in mental health may only be used for educational and administrative purposes. VanHook believes this framework aligns with those guidelines, offering a safe, supervised way to use AI for learning and clinical training. The research is published in the journal Frontiers in Health Services.
“AI is already reshaping the future of care,” VanHook said. “Our challenge now is to ensure it’s used responsibly, to teach, to train, and ultimately to help clinicians deliver more personalized, accessible, and effective mental health services.”
