Ensuring safe and responsible use of AI in healthcare
This project works with Australian health services to develop and implement an AI governance framework and AI maturity model to enable the safe and responsible use of AI in healthcare delivery.
Project sponsors: Digital Health CRC Limited, Bayside (Alfred) Health and Northern Sydney Local Health District
About the project
Artificial Intelligence (AI) technologies promise to transform care delivery, yet adoption in health service organisations remains slow. One of the main challenges for successful implementation and meaningful adoption of AI tools is their governance at the health service level. Such governance is critical not only to ensure safe and effective deployment, but to foster clinician trust that leads to meaningful adoption as well as improvements in care delivery and patient outcomes.
There is limited practical guidance for Australian health services about how to govern the safe and responsible implementation and use of AI. While there are a multitude of theoretical frameworks for AI ethics, there is little literature about how they are operationalised for the governance of AI applications in healthcare which are of varying levels of technical maturity, can incorporate many different types of computational reasoning methods including traditional and generative AI, and be used in a wide-variety of clinical and non-clinical areas.
In response to this gap, we have developed a practical AI governance framework to support health service organisations to:
- review
- approve
- implement
- and monitor the safe and responsible use of AI.
The framework translates high-level AI ethics and safety principles into operational governance processes, supported by an AI review checklist, and a maturity model that helps health services assess their readiness and capability to govern AI over time.
This project uses a mixed-methods design including literature reviews, document mapping, stakeholder interviews, case studies and workshops.
The overarching aims of this project are to:
- Develop, test and adapt an AI governance framework to provide oversight at the health service level and an AI maturity model to support the safe and responsible integration and use of AI tools in clinical and non-clinical areas.
- Apply and evaluate automated methods to monitor performance of AI tools over time to ensure they remain safe and effective.
Project outputs will be adopted by partners in New South Wales and Victoria to assist health services in navigating the complexities surrounding AI and ensure the safety of AI in healthcare, ultimately improving care delivery.
Project lead: Professor Farah Magrabi
Collaborative partners
Macquarie University has led this project in collaboration with:
- Bayside (Alfred) Health
- Amy McKimm
- Erica Potts
- Sudeep Safaf
- Digital Health CRC
- Judith Ngai
- Shiva Sharif
- Win Yee Tan
- Northern Sydney Local Health District
- Genevieve Baird
- Ann Mirapuri
- Karina Silva
- Andrew Wong
Together, the partners have developed an AI governance framework and AI maturity model to support the safe and responsible implementation of AI in Australian health services.