The Cutting-Edge AI Technologies Transforming Eye Care

AI is transforming the ways in which we look at the workplace.  In many sectors, heated debates rage on about the ethical usage of AI, with many ophthalmologists and optometrists concerned about the effect AI might have on the industry.  However, the IOC describes revolutionary AI machine learning technologies as the ‘fourth industrial revolution’ – something that we ought to embrace, learn about and adapt to. AI is proving to be an extremely useful tool in the ophthalmologist’s toolkit; it’s down to you to determine how best to use this emerging tech.  Of course, AI features are entirely optional in any instruments – if you do not wish to use them, you don’t have to, but the option is there should you need it at any time.  In this blog, we’ll cover some of the cutting-edge advancements we’ve seen in 2023, what the scientific community is saying about them, and what products are already using AI to great effect.  

What’s new in ophthalmology AI?

Conversations in the industry about AI are everywhere, so it’s impossible to condense the entire conversation into a single article, but we’ll run through some highlights here, starting with:  

The Optomed Aurora IQ’s Avenue AI software

Optomed, in many ways, are leading the charge in integrating convenient, powerful AI tools into their instruments.  The Aurora’s AI software improves the quality of, and accessibility to eye screenings, with Avenue AI being considered the ‘gateway to artificial intelligence services’.  With their cloud-based system, they’re able to integrate different AI algorithms into both cameras and screening software. Avenue is being used to extend diabetic retinopathy screenings to reach wider patient groups, as well as diagnosing other eye diseases such as suspected glaucoma or age-related macular degeneration (AMD). In addition to eye diseases, by analysing retinal images with AI, it will also be possible to detect signs and risk factors of various cardiovascular and neurological diseases.  

AI may help with estimating retinitis pigmentosa progression

Modern Retina reports on research conducted by Japanese ophthalmologists into how using AI alongside ultra-widefield fundus autofluorescence images may be helpful to objectively estimate progression of retinitis pigmentosa (RP) and visual function. Their goal was to ‘determine if AI can accurately estimate the visual function of patients with RP by using ultra-widefield fundus images obtained on concurrent visits.’ The team found that by using ultra-widefield fundus autofluorescence images alone provided the highest estimation accuracy for mean deviation, central sensitivity and visual acuity: ‘Standardized regression coefficients were 0.684 (95% confidence interval [CI], 0.567-0.802) for the mean deviation estimation, 0.697 (95% CI, 0.590-0.804) for the central sensitivity estimation, and 0.309 (95% CI, 0.187-0.430) for the visual acuity estimation (all p < 0.001).’

Fundus-based AI algorithms for ophthalmic and non-ophthalmic applications

The Artificial intelligence in ophthalmology 2023 report was published following the June 2023 conference, and holds many unique insights into how AI will shape our industry in the years to come.  One such insight focused on how fundus-based AI algorithms are being used for both ophthalmic and non-ophthalmic applications.  Studies showed that ‘The integration of AI algorithms for the interpretation of retinal images offers an inventive solution, exhibiting notably higher accuracy in the detection of ophthalmic and non-ophthalmic disorders in comparison to clinical experts.’ In short, AI allows for immediate analysis, allowing ophthalmologists and optometrists to process more patients in a shorter amount of time, all while enhancing precision, speed and workflow efficiency.  Furthermore, AI holds promise in minimising errors throughout the education and training process, providing trainees valuable tools to support them as they grow in experience and develop their knowledge.    

Predicting children’s myopia risk using machine learning models

The importance of early myopia detection is a topic that’s close to our hearts at Mainline, which means that reading about the incredible cutting-edge machine learning models that are being used to help with early detection is really exciting! A study, also published in the artificial intelligence in ophthalmology 2023 report, introduces a novel approach for predicting myopia risk through ‘trustworthy machine learning models developed by the Myopia Risk Calculator Consortium (MRC).’ Using data from just shy of 4,000 children, researchers focused on the trustworthiness of machine learning models, specifically how effective and robust their results are.  ‘The combination of classifiers and regression models yielded promising results for myopia risk prediction. Future research will explore incorporating image and synthetic data to enhance predictive accuracy through transfer-learning techniques.’ Any process that can help with early myopia detection and management is great news in our books, which is why we champion the use of the OCULUS Myopia Master® and its incredible auto-generative myopia management reports.   

This research sounds great, but what AI tech is available right now?

We understand if you’re apprehensive about AI – it’s natural; millions of people around the world in almost every industry are facing the prospect of AI and wondering how they’ll fit into the working landscape of the future.  However, with technology as it is today, AI serves as an incredible efficiency-boosting tool, one that will make your practice perform better, for your patients and for you. Here’s two key examples of some of the unique AI tech that’s already available:  

iCare DRSplus with iCare ILLUME

The iCare DRSplus is an incredible piece of kit with or without AI.  Its Confocal Fundus Imaging system uses white LED illumination to produce TrueColor and detail-rich images, setting a new standard of retinal imaging.  iCare ILLUME, a specialised screening AI software, helps detect the early signs of diabetic retinopathy, age-related macular degeneration, and glaucoma using artificial intelligence. This:

• Reliably and efficiently reviews images
• Saves your valuable time
• Enables early intervention
• Helps optimise your workflows

For the first time, screening is possible for all patients, including those with ocular media opacities such as cataracts and vitreous cloudiness.  All AI is cloud-based. It keeps patient data anonymised; data is never shared and patient confidentiality is a core pillar on which the ILLUME AI system was developed.   

Optomed Aurora IQ

We mentioned Optomed’s Aurora IQ Handheld Fundus Camera a little earlier, but we’ll go into a little more detail here. The Optomed Aurora IQ is a handheld fundus camera that has been designed to meet your needs both in usability and image quality. The camera is made for examination and documentation of the retina and anterior of the eye for various eye diseases and neurological disorders. It easily travels to the patient or with you on-call for maximal workflow efficiency. Its AI software, Avenue, improves the quality and accessibility to eye screenings, all while:

• Integrating various AI algorithms with their cameras and screening software
• Extending diabetic retinopathy screenings
• Diagnosing other eye diseases such as suspected glaucoma

It’s an incredible tool, designed to supplement your expertise and help you make decisions faster than ever before.  

Need advice on how AI tools might integrate into your practice’s workflow?

Then don’t hesitate to talk to our team at Mainline Instruments.  It’s our job to connect optometrists and ophthalmologists with the best equipment for their practices and patients.  If you’d like to learn more about anything discussed in this blog, or you’ve got a specific product you’re interested in hearing more about, you can get in touch with us by clicking the link below. Talk to our equipment specialists Prefer to browse our products instead? Click here to look through our extensive catalogue of products.