Business, Company, Local, News, Ophthalmology, Optometry, Research, Technology

Multi-million dollar Aussie project to cut optometry false-positive rate

Australian agencies are collaborating with technology firm Big Picture Medical in a $12 million project to develop an AI-assisted platform that could reduce inappropriate patient management at the primary point-of-care.

Led by Big Picture, the venture will involve intellectual property from Sydney and Adelaide-based research institutions that will be integrated into the company’s platform that combines artificial intelligence (AI), data analytics and decision support algorithms to drive better patient outcomes.

If successful, Professor Fiona Stapleton, one of lead researchers in this project, said the collaboration would provide optometrists with a platform that could analyse multiple patient eye scans, as well as medical history data with significantly improved diagnostic accuracy.

Fiona Stapleton, lead researcher.

Big Picture was recently awarded a $2.9 million government grant for this project under the Cooperative Research Centre Project Grants program, which supports short-term collaborative research projects.

The collaboration will also involve the University of Adelaide’s Australian Institute of Machine Learning, Brien Holden Vision Institute (BHVI), Menicon Australia and the University of New South Wales (UNSW), comprising its School of Optometry and Vision Science, Centre for Eye Health (CFEH) and School of Computer Science and Engineering.

CFEH director Professor Michael Kalloniatis told Insight that Big Picture, which is based in both Sydney and London, has an AI-powered software platform that “structures and liberates medical dark data” to improve patient outcomes. In eye health, the platform has already linked primary eyecare practitioners to other practitioners such as GPs, as well as hospital ophthalmology departments.

The company had expressed a desire to incorporate AI into the platform to help the referring practitioner correctly manage patients with eye disease.

“Most AI systems only consider raw clinical images – a unique feature of this project is that the platform will combine patient data such as age and other risk factors with the scans and provide a superior algorithm to help guide primary care clinicians in patient management,” he said.

“The key point is that this will guide optometrists, it will not make decisions for them.”

In Australia, misdiagnosis can be relatively common in optometry, leading to poor patient outcomes and unproductivity for limited specialist services. This results in an estimated $74 million per annum in avoidable costs for the Australian health system.

Kalloniatis said Big Picture had previously worked with Moorfields Eye Hospital in London to examine referrals to hospital ophthalmology departments, which showed high false-positive rates – 37% in the National Health Service (NHS) – for optometric referrals.

Michael Kalloniatis, CFEH.

“The average number of visits in cases of false positive referrals is around 2.2, which reflects wasted public clinical resources. Big Picture’s platform significantly reduced the false-positive rates – effectively halving it – but there are still too many inappropriate referrals,” he said.

“The expectation is that the AI-assisted platform will further reduce unnecessary referrals and improve patient outcomes and also improve communications between all health care providers.”

According to Kalloniatis, the NHS has parallels to Australia, which could translate into a similar reduction in false-positive referrals currently experienced on public hospital waiting lists.

What each entity brings to the table

  • Big Picture is the lead commercial partner with an AI powered software platform that structures and liberates medical dark data to transform patient outcomes, facilitates commercial insights and changes the delivery of healthcare for the benefit of all.
  • The UNSW’s School of Optometry and Vision Science and CFEH have the clinical expertise, as well as being in a position to test the AI algorithms in a practical setting.
  • The UNSW’s School of Computer Science and Engineering and University of Adelaide’s Australian Institute of Machine Learning have the expertise in AI algorithm development.
  • BHVI has access to remote and international communities to test the program.
  • Menicon has a history of working with the UNSW’s School of Optometry and Vision Science and has relationships with eyecare practitioners internationally.
  • Previously, Big Picture and CFEH have teamed up in a number of small projects and the School of Optometry and Vision Science has collaborated with Menicon, providing expertise in eyecare.