AI to benefit youth mental health

Artificial intelligence (AI) is set to benefit the mental health of young Australians, thanks to an almost $5 million investment from the Commonwealth Government’s Medical Research Future Fund (MRFF) as part of its digital health intelligence priority stream.

The program will be led by Scientia Professor Helen Christensen (AO) (Director and Chief Scientist, Black Dog Institute) and ARC Laureate Fellow and Alfred Deakin Professor Svetha Venkatesh (Co-Director of the Applied Artificial Intelligence Institute (A²I²)).

The other A²I² Chief Investigators are Associate Professor Sunil Gupta, Associate Professor Santu Rana, Associate Professor Truyen Tran, Dr Thomas Quinn,  Professor Kon Mouzakis (Co-Director), and Professor Rajesh Vasa.

In this innovative three-year project, researchers from the Black Dog Institute and Deakin University will use AI to determine the most effective psychological therapies for university students experiencing psychological distress. The project will explore whether some of the therapies are more effective in certain groups and uncover the characteristics of such groups.  Once proven successful, it is likely to open the door to improved treatment for many people experiencing mental health conditions in Australia and overseas.

“The AI developed for this project can be used in other significant health challenges where time and cost in identifying which care strategies work best are an issue. Other health challenges that stand to benefit include the management of diabetes, neurological conditions, addiction, and medications, sleep treatment, and primary care health promotion,” Professor Venkatesh said.

Scientia Professor Helen Christensen AO said the Black Dog Institute was grateful to have secured the project funding.

“Most psychological interventions for mental health problems are only partially effective. Artificial intelligence techniques can be used to help determine which of a range of therapies is most effective and for whom. AI can also be used to personalise treatments to individual characteristics,” Professor Christensen said.

Adaptive experimental design uses machine learning and lean data to achieve sample efficient trials, and thus has potential to reduce cost. This approach has proven successful in several domains from manufacturing to health, including a 2019 project with 26 Geelong GPs, where results are published in Nature Partner Journals (NPJ) Digital Medicine.