Australian optometrists now have access to ‘Eyetelligence Workstation’ – a user friendly, fast and accurate artificial intelligence (AI) platform that detects four most common eye diseases: diabetic retinopathy, glaucoma, age-related macular degeneration and severe cataract, in a matter of seconds. The platform has been described as providing diagnostic accuracy, clinical decision support, and clinical productivity.
Eyetelligence Workstation was developed by the Australian AI company Eyetelligence, and recently received Therapeutic Goods Administration (TGA) approval to offer its ‘software as a medical device’ platform.
Dr Harrison Weisinger, Former Director of Medical Affairs for Specsavers, who was recently appointed Commercial Director for Eyetelligence, said unlike many AI technologies, Eyetelligence Workstation has been developed for use in ‘real world’ practice.
Eyetelligence Workstation is a unique product by way of the fact its algorithm was ‘trained’ using images from 70,000 patients…
“A number of research groups and companies in the world are working on artificial intelligence technology based on fundus photography but very few, if any, have developed a product that is suitable for real world optometry practice. Typically, AI products demand massive computing power which means the users have to upload the images into an internet cloudbased platform or update their hardware to expensive, high-spec machines. The impact on clinic flow, as well as privacy concerns, basically makes such cloud-based alternatives a nonstarter,” said Dr Weisinger.
“The wonder of Eyetelligence is that it runs locally. Deployment of an AI product in realworld practice is challenging, not only demanding accuracy in the detection, but is also dependent on how easy it is to use, how fast the detection is, how it deals with poor image quality and more importantly how it interacts with the clinicians and patients”.
According to Eyetelligence’s Chief Medical Officer and clinician-scientist from University of Melbourne and Centre for Eye Research Australia (CERA), Professor Ming He, “Eyetelligence Workstation is a unique product by way of the fact its algorithm was ‘trained’ using images from 70,000 patients and validated in multiple ethnic groups including Caucasians, Chinese and Indigenous Australians.
“The validation results have been published in numerous world class journals in the fields of ophthalmology and diabetes. The system has been developed and refined over a period of four years for local clinical use. It operates 100% locally in the clinic, does not require image upload via internet and every patient’s identity, information and clinical data are protected. This system not only detects four most common eye diseases, such as diabetic retinopathy, glaucoma, age-related macular degeneration and severe cataract, in a matter of seconds, it also automatically identifies right and left eyes, assessing if the images are acceptable for grading and more importantly the system works silently at the backend without need for additional operations by users.”
Professor He is acknowledged to be a leader in his field. This month, he was awarded a record three-year, AU$5M grant by the Australian Government’s Medical Research Future Fund (MRFF) to lead the development of an integrated AI eye screening system for use in primary care settings such as GP clinics, Indigenous health services, eye care clinics, endocrinology and cardiology services to increase access to life and sight-saving screening services. Eyetelligence was named in this grant as the exclusive technical and commercial partner.
Optometrist and practice owner, Stephanie Park, has used the Eyetelligence Workstation in her practice for seven months.
“The beautiful thing about the system is that within seconds of taking a patient’s fundus photograph, the optometrists in my practice, and I, can be almost completely certain of whether that person does or doesn’t have a major eye disease – and this is even before you’ve asked them a single question or examined their eyes. The advantages for diagnostic accuracy, clinical decision support and even clinical productivity are enormous”.
Eyetelligence currently works with several popular imaging devices, but the company has plans to expand its offering to work with all major device suppliers. It also intends to release several new features, including optical coherence tomography diagnostics as well as non-eye related attributes including biological age, Alzheimer’s disease risk and cardiovascular disease risk.
Optometrists who are interested in trialling the Eyetelligence Workstation should contact Dr Weisinger: email@example.com.