Australia ‘unprepared’ to handle risks of AI-powered health system

While Australia is capable of becoming a global leader in the application of AI in healthcare, it still has heaps of work to do to harness the full benefits of the technology.

Its healthcare AI market has been projected to grow to $1.78 billion by 2030 from $80 million last year. AI holds the key to a “smarter, more adaptive health system” in Australia that can withstand emerging demands such as increasing disease complexity, treatment intensity, and future pandemics.

The country may have the necessary elements to apply AI in healthcare, including the digital infrastructure, decades of research, and a high-functioning health system, but it is “currently unprepared” to handle the opportunities and risks of an AI-enabled health system, according to the Australian Alliance for AI in Healthcare (AAAiH).

For one, its current model for AI regulation was designed over 30 years ago when technology was “single function and predictable.” 

Its existing healthcare AI industry is relatively small compared to other developed nations: the United States, for example, has so far granted clearance to more than 500 new AI-driven medical devices, while the British government has invested as much as £2.3 billion ($2.9 billion) in AI initiatives over the past years, including £250 million ($315 million) to accelerate the AI adoption in the National Health Service. 

Australia is also lagging behind the research investment for healthcare AI; its overall healthcare AI research capability lacks global competitiveness and its R&D support is nowhere near those made available in other nations. Over the past five years, not a single clinical trial study involving AI was conducted in the country while local AI research only attracted 0.9% of total government grant funding. 

Given these limitations, the AAAiH has developed the National Policy Roadmap for AI in Healthcare intending to lift Australia to the bar of other nations that have already made significant strides in investment and adoption of healthcare AI. 

The first version of the roadmap was released two years ago. Since then, technological developments, including the introduction of generative AI, prompted the alliance to convene stakeholders once more to update it. Laying the foundations for reaping AI’s full benefits, the new roadmap offers 16 recommendations across five priority areas:

• AI Safety, Quality, Ethics and Security
• Workforce
• Consumer
• Industry
• Research

“AI offers us profound new opportunities to improve clinical diagnosis, treatment and workflows. From research bench to clinical bedside and into the hands of patients, AI promises to make Australian healthcare a learning system that is more nimble, adaptive, personalised, safe, effective and equitable,” the roadmap emphasised.

The roadmap’s development was supported by Macquarie University, the CSIRO Australian eHealth Research Centre, RMIT University, the Digital Health Cooperative Research Centre, and the Australasian Institute of Digital Health.

ON THE RECORD

Among pressing areas of healthcare AI, governance must be high up in the federal government’s priorities, said Enrico Coiera, Macquarie University professor and founder of the AAAiH. Governance, it is said, must move from a “certify once” model to “one which ensures adaptive AI remains fit for purpose as it evolves.”

“Australia needs to move fast to safeguard patients and support our healthcare and AI sectors while taking advantage of the benefits and mitigating the risks of AI,” he said in a media release.