AI applications for mental health

Software	None
Covers	Human factors, artificial intelligence
Skills Required	Interest in AI and autonomous systems, interest in mental health issues
Challenge	Conceptual ???? Technical ??? Programming

Brief Description

How can Artificial Intelligence technology be applied to monitoring, treating or preventing mental health issues?

The challenge of this project is to look at how AI technology (or particular types of AI technology, such as machine learning algorithms for monitoring patient behaviour, or autonomous robot pets, or expert systems for diagnosis, or tutoring systems that enable actual or potential sufferers to educate themselves about their conditions) can help sufferers of mental illnesses. For example, dementia patients might gain from having robot pets they can interact with without needing to do the difficult parts of looking after a living creature, by getting company, mental stimulation to keep them more alert, or having a system there to monitor their behaviour and alert carers to abnormal situations like falls.

A project on this topic is likely to be most successful if it focuses on a particular issue or type of AI technology or mental health problem.

What are the advantages and disadvantages of using computer technology including AI for monitoring patients with mental illnesses? How do these relate to the limits of what you can do with AI technology? (Which technologies for which patients?)
What kinds of interactions with robots or other autonomous systems might help sufferers or harm them?
How might people with particular medical conditions develop different mental models of robots or other AI systems? Would this be helpful or harmful, and if so how?
Are the systems and techniques under consideration potentially useful for some people who qualify as 'normal' or 'healthy'?

Variants

A rather different project would be picking an idea for a system to help a sufferer from some form of mental illness or disability, and developing a concept demonstrator prototype.

This could be a computer ethics project: What are the ethical implications of using systems with particular characteristics with particular kinds of patients? For example: using robots to alter mood and behaviour; AI behaviour monitoring to which people can't give full consent.

Back to