Artificial intelligence to improve optometrists’ diagnoses and referrals 

Researchers and industry are collaborating to develop AI capabilities that analyse multiple patient eye scans and medical history data with specialist-level accuracy at the primary point-of-care.

The Centre for Eye Health (CFEH) - a joint initiative of Guide Dogs NSW/ACT and UNSW Sydney - and the Schools of Optometry and Vision Science and Computer Science and Engineering at UNSW, are partnering with global technology company Big Picture Medical (BPM) on a Cooperative Research Centre Project (CRC-P) that will reduce the number of specialist consults and save considerable avoidable costs within the Australian health system. Also partnering on the project are the Australian Institute of Machine Learning at the University of Adelaide, Menicon Pty Ltd and the Brien Holden Vision Institute.  The technology will combine an optometrist’s expertise with AI capabilities that analyse comprehensive patient eye health data to achieve specialist-level accuracy to better diagnose and manage eye disease.

Federal CRC-P grants support short term industry-led collaborations to develop important new technologies, products and services that deliver tangible outcomes to the community.

Professor Michael Kalloniatis, Director of CFEH, says misdiagnosis at the primary point-of-care is a well-established problem that results in poor patient outcomes and unproductive use of limited specialist services.

 “AI algorithms can provide guidance to the clinicians particularly in early diagnosis of eye disease, thereby reducing the significant false positive referrals to public hospital eye departments and also help detect disease early before irreversible vision loss occurs”, says Professor Kalloniatis

Researchers and industry peak bodies, including the Royal ANZ College of Ophthalmologists (RANZCO) and Optometry Australia (OA), have identified that artificial intelligence (AI) can improve the accuracy of diagnosis and lead to better patient outcomes and is likely to play a major role in eye care in the future.

To be useful in practice, AI systems need to be trained to consider and analyse broad patient data including multiple images, vision tests and historical clinical data. The industry’s adoption of AI has also been hampered by concerns about automation and the comprehensiveness of AI analysis. While optometrists are strong adopters of technology and recognise AI’s potential, instead of full automation, they are looking for solutions that incorporate their expertise into final analyses. To give them confidence in the recommendations, they also want AI systems with transparent methods and rationale.

Big Picture has developed a world leading technology platform that combines advanced AI, data analytics and decision support algorithms to deliver improved and more accessible patient outcomes.  BPM’s software can be utilised across all medical categories, and provides the foundation for an AI system that can consider multiple sources of data. The CRC-P will incorporate unique intellectual property (IP) from UNSW’s Schools of Optometry and Vision Science and Computer Science and Engineering, the CFEH and the University of Adelaide’s AIML into Big Picture Medical’s platform.

Big Picture Founder & CEO, Dr Tom McKinnon said “We thank the Australian Government for this grant. Our mantra is to facilitate the improvement and accessibility of healthcare for all. AI has the opportunity to play an important role in delivering that aspiration. We have built the software platform that enables AI developers and health companies from all over the world to apply their applications to deliver improved and more accessible patient outcomes. This grant will help accelerate that process for the benefit of patients everywhere.”

A key activity of the project includes training algorithms to read, map and analyse diverse patient data to develop a diagnostic recommendation. The project team will work with optometrists in urban, rural and regional Australia to optimise the interface. The system’s performance and usability will be validated through prospective trials within UNSW facilities, CFEH clinics in Australia and through partnerships in China facilitated by Menicon.

 

For further information, please contact:

Scientia Professor Fiona Stapleton: (02) 9385 4375 or f.stapleton@unsw.edu.au

Professor Michael Kalloniatis: (02) 8115 0710 or mkalloniatis@cfeh.com.au