Implementing artificial intelligence (AI) to enhance Lifeline’s crisis support service capacity in response to COVID-19 and emerging crises

  • Rickwood, Debra (CI)
  • Goecke, Roland (CoI)
  • Larsen, Mark (CoI)
  • Epps, Julien (CoI)
  • Klein, Britt (CoI)
  • Dhall, Abhinav (CoI)
  • Ma, Jennifer (CoI)

Project: Research

Project Details


This research seeks to address the mental health and wellbeing needs of Australians by implementing artificial intelligence (AI) to enhance Lifeline’s crisis support service capacity in responding to the short and long-term negative impacts of the COVID-19 pandemic and future crises.

The project aims to rapidly develop, implement, and evaluate the use of artificial intelligence to enhance the effectiveness and responsiveness of the Lifeline crisis support service in managing the mental health and wellbeing community impacts of crisis, including COVID-19. Machine learning algorithms will be developed to better:
1. Efficiency - Identify and direct help-seekers based on presenting crises and characteristics.
2. Effectiveness - Assess help-seeker outcomes via real-time monitoring of speech and text valence.
3. Practice support - Co-create a visualisation engine that summarises topic distributions in real time, and the extent to which help-seekers are achieving positive outcomes

The work will be carried out primarily at the University of Canberra. Most of the project involves the analysis of existing data, so there are no research sites. The research design comprises coding and analysis of existing data held by Lifeline to develop AI algorithms to determine caller issues and characteristics and outcomes. There is also a small qualitative component to gain end-user input to the development of practice supports for Lifeline.

This proposal builds on a current NHMRC partnership grant held by CIs Rickwood and Goecke. The proposal aims to obtain additional resources to better undertake and extend the current NHMRC grant.
Effective start/end date1/11/2030/04/22


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.