Some individuals with mental health disorders hesitate to discuss personal issues with a clinician due to concerns about judgment or stigma. This is especially true for people with addiction disorders, yet foregoing professional help can harm both mental and physical health. Now, researchers at Cedars-Sinai say artificial intelligence (AI) may play a role in addressing these concerns.
Two new Cedars-Sinai studies indicate that AI can help deliver mental health therapy. One study focused on AI-driven avatars that conducted therapy sessions with patients managing alcohol addiction and received positive feedback. A second study found these virtual therapists offered unbiased counseling regardless of a patient’s race, gender, or income.
Cedars-Sinai investigators trained an artificial intelligence application to offer mental health therapy and provide unbiased counseling regardless of a patient’s race, gender, income, or other traits. Credit: Getty
Both studies used a Cedars-Sinai–developed system that combines AI and virtual reality (VR) goggles. The application features avatars “trained” by AI to deliver talk therapy in virtual, calming environments.
In the first study, published in the Journal of Medical Extended Reality, 20 patients with alcohol-associated cirrhosis participated in a 30-minute virtual counseling session guided by an AI-driven avatar. More than 85% said they found the sessions beneficial, and 90% expressed interest in using virtual therapists again.
Brennan Spiegel, MD, MSHS Credit: Cedars-Sinai
“For individuals awaiting liver transplants for cirrhosis, alcohol addiction remains a high-risk factor,” said Brennan Spiegel, MD, MSHS, professor of Medicine, director of Health Services Research at Cedars-Sinai, and corresponding author on both studies. “We see VR as a way to augment traditional interventions, which often fall short due to a shortage of mental health professionals, societal stigmatizing of alcoholism, and other factors.” Spiegel was also involved in recent research showing that VR experiences could affect participants’ stress levels and immune responses.
The second study, published in Cyberpsychology, Behavior, and Social Networking, paired virtual therapists with virtual patients whose AI had been trained to reflect people seeking help for anxiety or depression. In each simulated conversation, the therapist was randomly given a patient profile based on factors such as age, gender, and income. Some conversations did not include a profile at all. Repeated samplings of more than 400 conversations found no difference in the therapist’s “tone score” across different patient profiles. “This data suggests that with thoughtful design, AI can offer equitable and personalized care,” Spiegel said.
Peter Chen, MD, professor of Medicine, the Medallion Chair in Molecular Medicine, and interim chair of the Cedars-Sinai Department of Medicine, noted these findings show Cedars-Sinai’s commitment to exploring AI while ensuring it does not reinforce human biases. “These two studies underscore Cedars-Sinai’s commitment to exploring the tremendous potential of artificial intelligence for mental health therapy while ensuring that this technology does not perpetuate human biases in delivering healthcare,” Chen said.
R&D World covered AI’s use in mental health care in 2020, with an article about AI being used to streamline the coding of national suicide-related ambulance data to help paramedics respond more effectively and ultimately prevent potential suicide rates.
Additional Details and Disclosures
Authors of the Journal of Medical Extended Reality study include: Yee Hui Yeo, Allistair Clark, Muskaan Mehra, Itai Danovitch, Ju Dong Yang, Alexander Kuo, Hyun-Seok Kim, Aarshi Vipani, Yun Wang, Walid Ayoub, Hirsh Trivedi, Jamil S. Samaan, and Omer Liran. Collaborators also included Karen Osilla, Tiffany Wu, and Vijay H. Shah.
For more information about academic research from Cedars-Sinai, visit Cedars-Sinai Academic Medicine on LinkedIn.
AI’s potential in therapy is fascinating! It could help reduce stigma and make mental health support more accessible. I’m curious about how it can be integrated into traditional methods without losing the human touch.