Using Artificial Intelligence to prioritise emergency calls for suicide prevention due to COVID-19 pandemic.
The COVID-19 pandemic has not only taken a toll on those who have contracted the virus but also had an impact on the mental health and well-being of Queenslanders – and communities across the globe.
So, how can we support the mental health and well-being of communities during a pandemic? Well, that’s exactly what University of Southern Queensland researcher, Dr Rajib Rana plans to address.
As one of the 2020 Advance Queensland Research Fellows, Dr Rana will research how Artificial Intelligence can be used to prioritise emergency calls for suicide prevention.
We caught up with Rajib to find out how this research could help those in distress during a pandemic such as COVID-19.
Could you tell us about your research?
My overarching research area is affective computing and I am currently focusing on creating highly accurate automatic emotion recognition techniques by developing state-of-the-art Artificial Intelligence and Machine Learning algorithms.
With this fellowship, I hope to determine distress levels in real-time, during a call to the emergency helpline.
My research is supported by scientific evidence that shows involuntary physiological responses to distress cause:
- rapid respiration rates
- a decrease in saliva production
- laryngeal muscle tension.
This results in extreme changes in speech production, which is not typically produced or required for day-to-day interactions.
I also hope to discover the speech features that can accurately describe distress in speech and propose a deep learning neural network framework that can learn distress representation from limited labelled data.
The expected outcome would be a set of computer algorithms underpinning a novel deep learning neural network architecture that takes a speech sample as input and instantly calculates the distress level.
How will the funding from Advance Queensland support you on your research journey?
Without funding support from Advance Queensland, I would not be able to carry out this vital work. This funding has allowed me to have the necessary research workload allocation and given me access to the essential research personnel and other resources that are crucial to carrying out the research.
How will your research benefit and assist Queenslanders?
In Australia, 25.7% of all deaths are by suicide. With Queensland feared to reach a historically high unemployment rate after the COVID-19 pandemic, mental health experts predict this could bring devastating mental health impact, drastically increasing the suicide rate – which is already very high in Queensland.
A heightened level of distress has a strong association with reports of suicide ideation and attempts. The proposed distress inference system implemented in the 1300 MH CALL service will assist clinicians in using distress levels of callers to identify hoax calls and, most importantly, use it to prioritise services to the callers who are most at risk, thus saving more Queensland lives.
It will also help the 1300 MH CALL service to effectively manage the high volume of calls received during and after COVID-19, by using two-tier approach. This means an automated system would first receive a call, and then based on the distress level present, a prompt, to refer the call to a second tiered response if required - helping those in distress, more efficiently.
The project will also help our partner Netcare expand its business in Queensland, creating many jobs for Queenslanders in the future. This world-first innovation also has strong global relevance and will provide tremendous commercial opportunities for Queensland technology.
The strength of this project is it will not only help prepare for future health crises in Queensland, but also help prepare for us for other future crises already known to result in high distress to Queenslanders, such as bushfires, droughts and floods.
What does it mean to you to be a recipient of the Industry Research Fellowships program?
The fellowship will allow me to develop the world’s first technology to determine distress automatically, providing me the opportunity to further develop my research leadership in the field of affective computing and gain reputation nationally and internationally. This will accelerate my career growth as a researcher and, most importantly, will be a massive encouragement to remain in Queensland.