UWA researchers to predict suicide risk with artificial intelligence

Researchers from The University of Western Australia are developing a system using artificial intelligence that can predict the likelihood of suicide in individuals most at risk, and assist with clinical decision-making.

The research team from UWA and the Mental Health Service at Armadale Kalamunda Group has been awarded a Mental Health Research Fund Grant by the East Metropolitan Health Service to create a system that will improve the existing risk assessment framework used by healthcare providers.

The current methodology to predict an individual’s risk of suicide or self-harm is highly subjective and can be influenced by the assessor’s experience, recent negative outcomes and other biases.

Through the development of a computer-based system using artificial intelligence algorithms, the researchers will be able to determine the factors in actual behaviour which are better predictors of risk.

The system would be developed as a companion tool to be used alongside the current risk assessment models and protocols.

Professor Mohammed Bennamoun from UWA’s Department of Computer Science and Software Engineering said a system that could successfully improve prediction of self-harm or suicide would have worldwide implications.

“It will also be particularly useful in places where there are fewer mental health specialists available to carry out the risk analysis,” Professor Bennamoun said.

Dr Dharjinder Rooprai, Head of Psychiatry and Consultant Psychiatrist at the Armadale Kalamunda Group, said a high proportion of individuals who went on to commit suicide often suffered with a mental illness, including depression.

“An artificial intelligence system that can be employed by healthcare providers to assist their patients’ risk of suicide or self-harm assessment would have a profoundly positive impact on their patients’ care journey,” Dr Rooprai said.

The researchers are also considering facial recognition as a tool to identify an individual’s mental state and predict the likelihood of adverse outcomes.

Professor Bennamoun said that while, for example, facial recognition was increasingly used for tracking and surveillance, artificial intelligence overall would have a significant impact on humans in new ways.

“The capability of facial recognition to identify patients at risk of suicide would be revolutionary in the field of psychiatry and would have implications far beyond risk prediction,” he said.

/University Release. The material in this public release comes from the originating organization and may be of a point-in-time nature, edited for clarity, style and length. View in full here.