Ophthalmology is at the forefront of the artificial intelligence (AI) revolution in healthcare.

Powerful imaging technologies combined with deep learning algorithms are transforming our ability to screen for eye diseases, with some new tools now able to analyse images of the retina at least as well as doctors.

A new report published by the Australian Academy of Health and Medical Sciences (AAHMS) is now calling for a national conversation about the opportunities and challenges that AI technologies bring to the future of Australian healthcare.

CERA Deputy Directors Professor Robyn Guymer and Associate Professor Peter van Wijngaarden were among 34 experts across health, technology, academia, industry and government who took part in a roundtable meeting exploring these issues.

The resulting report, Artificial Intelligence in Health: Exploring the Opportunities and Challenges, concludes that if we are to successfully harness the potential value of intelligent health technologies, we need to determine a national pathway for its ethical and responsible use, and how the current health care model will need to adapt to technological changes and their impact on healthcare delivery.

Artificial intelligence in eye care

At CERA, AI is playing an increasingly significant role in research innovation. CERA researchers are demonstrating that AI tools have the potential to make eye screening more accessible and convenient for patients, and to help doctors predict the progression of eye disease.

“New technologies have the potential to detect the early signs of disease – without the need to attend an eye clinic – and determine who needs to be referred to an eye care professional for a more detailed assessment,” says Associate Professor van Wijngaarden.

“Whilst not yet in mainstream use, research studies are revealing the potential to test patients at locations that suit them – in their own home, a photobooth in a shopping centre or during a visit to another health care provider, such as a GP.’’

In ophthalmology and across the healthcare industry, judicious use of AI technologies could potentially allow for more cost-effective care, reduce stressful workloads for clinicians, and provide more flexibility and empowerment for patients.

Challenges to consider for the future

While AI technologies offer incredible potential, the AAHMS report also reveals the complexity of this emerging landscape.

This includes challenges around ensuring validity, safety, ethics and accountability for decision-making.

“Optimism about the potential of AI to enhance access to care and convenience for patients needs to be tempered by the limitations of the technology and the readiness of patients and health systems for its use,” says Associate Professor van Wijngaarden.

“Acceptance of AI by patients and care providers, its performance in real world clinical settings and its place alongside medical professionals all need to be evaluated.”

Associate Professor van Wijngaarden says a broad range of perspectives need to be considered to determine a pathway.

“The report highlights that it is high time to initiate a national conversation about how best we should use AI in healthcare to ensure that we are optimally placed to benefit from the safe and appropriate use of this technology.

“Broad-based consultation involving consumers, regulators, researchers, care providers, technology developers and governments are needed.”

Read the full report, Artificial Intelligence in Health: Exploring the Opportunities and Challenges, from the Australian Academy of Health and Medical Sciences.