Mental health disorders are one of the most serious diseases in the world. Most people with such a disease lack access to adequate care, which highlights the importance of training models for the diagnosis and treatment of mental health disorders. However, in the mental health domain, privacy concerns limit the accessibility of personalized treatment data, making it challenging to build powerful models. In this paper, we introduce MentalArena, a self-play framework to train language models by generating domain-specific personalized data, where we obtain a better model capable of making a personalized diagnosis and treatment (as a therapist) and providing information (as a patient). To accurately model human-like mental health patients, we devise Symptom Encoder, which simulates a real patient from both cognition and behavior perspectives. To address intent bias during patient-therapist interactions, we propose Symptom Decoder to compare diagnosed symptoms with encoded symptoms, and dynamically manage the dialogue between patient and therapist according to the identified deviations. We evaluated MentalArena against 6 benchmarks, including biomedicalQA and mental health tasks, compared to 6 advanced models. Our models, fine-tuned on both GPT-3.5 and Llama-3-8b, significantly outperform their counterparts, including GPT-4o. We hope that our work can inspire future research on personalized care. Code is available in https://github.com/Scarelette/MentalArena/tree/main

There are a plethora of AI initiatives in progress across the healthcare industry. From drug discovery to thermal scans, AI has transformed this sector significantly over the past decade. While we know AI's contributions to physical healthcare, AI is also easing mental health concerns too. According to the Substance Abuse and Mental Health Service Administration (SAMHSA)'s 2016 report on drug use and health, only 63% of adults identified as having had at least one major depressive episode reported receiving any kind of treatment. In the US alone, one in five adults suffers from a form of mental illness.

A team from Huddersfield has pioneered the use of artificial intelligence to predict those mental health patients most likely to take their own lives. Prof Grigoris Antoniou and a team at the University of Huddersfield worked with South West Yorkshire Partnership NHS Foundation Trust (SWYPFT) to examine how AI could help reduce the risk of suicide among mental health patients. Now further work will be carried out into ways of deploying the new computer-based technique. Although clinical decisions about suicide risk will always be based on clinical judgement, findings from this research should aid decision-making by identifying high risk patients based on learning from previous suicides. Prof Antoniou, a globally-acknowledged expert in AI technologies, and his team analysed data from more than 100 suicide cases in order to compile a list of the key risk factors.