AI tech to bolster suicide prevention efforts in Australia

Monash University and Eastern Health's Turning Point are using a $AUD1.2 million grant from Google to address major mental health issues in society.

A joint project between Monash University and Eastern Healths Turning Point sees the use of artificial intelligence (AI) to streamline the coding of national suicide-related ambulance data to help paramedics respond more effectively and ultimately prevent potential suicide rates.

Ambulances are often the first point of contact in a crisis, making ambulance clinical records a unique data source to help inform suicide prevention efforts.

In partnership with the population health team at Turning Point, Monash Faculty of IT researchers are implementing specialised machine-learning technologies to filter through ambulance data and categorise suicide-related mental health cases.

As the only Australian grant recipients of the 2019 Google AI Impact Challenge, the project will annotate a large database of clinical records into categories such as suicide attempt, suicidal ideation and self-injury without suicidal intent.

This task is currently carried out by human annotators who interpret individual paramedic clinical records and categorise it accordingly.

By using the data that has been previously annotated, the machine-learning model will be able to augment and accelerate the challenging work of categorising and filtering through tens of thousands of records per month.

A key outcome of the project has been ensuring greater efficiency and accuracy in data classification.

With preliminary studies showing that two-thirds of the data processing can be eliminated, potentially reducing staff workload by over 30%, thereby freeing them up for other tasks and limiting their exposure to explicit content.

Implementing this machine learning model will result in timely and cost-effective identification coding of suicide-related ambulance data, to better inform policy and public health responses for suicide prevention.

By accurately classifying data records, researchers will be able to determine behaviours and relationships associated with certain classifications.

For example, how self-harm incidents can relate to other factors such as violence, drug use and socioeconomic status.

"The coaching and support we've received through the Google AI Impact Challenge has been invaluable and amplified our projects overall impact," says Monash University professor of data science and AI project lead Wray Buntine.

"The interactive sessions we've had with members of Googles Emerging Tech Design team have allowed us to work with international leaders across the fields of people-centred AI.

"Through our shared values, we're committed to using IT for social good," says Buntine.

By leveraging Googles expertise and tools, researchers have been able to explore new ways of confronting a major societal issue.

Google program manager Mollie Javerbaum says the project is a great example of how emerging AI technology can be leveraged to create positive social impact while advancing the AI ecosystem.

"The Monash University and Turning Point teams have continued to make meaningful strides towards improving their suicide monitoring system through the application of AI," Javerbaum says.

Turning Point is a national addiction treatment centre, providing evidence-based treatment to people adversely affected by alcohol, drugs and gambling.

Turning Point director and Monash University professor of addiction studies and services Dan Lubman believes the initiative has the potential to set international standards in supporting suicide prevention efforts.

"The project will uncover critical suicide trends and potential points of intervention to better inform policy and public health responses," says Lubman.

"The technology we're developing will create opportunities for adoption internationally.