Early detection of ocular pathologies with AI

There is a popular saying that goes: “prevention is better than a cure”. In medicine, the great challenge is precisely to prevent diseases or to diagnose them as early as possible.

In ophthalmology, the prevention of some diseases is becoming more and more important. A minimal eye examination for visual acuity, pupil function, and extraocular muscle motility, is capable of detecting changes such as refractive error amblyopia, diabetic retinopathy, glaucomatous optic neuropathy, age-related macular degeneration, and more.

Why is Artificial Intelligence spreading to ophthalmology

In the past years, there has been a major progress both in theoretical research and applications of Artificial Intelligence (AI), especially in the field of Deep Learning (DL). Thanks to these advances, today AI can help advance early detection of ophthalmology diseases with a high occurrence rate, such as diabetic retinopathy, glaucoma, age-related macular degeneration, etc. - based on patients’ eye images.

A particularly exciting development in the field of ophthalmology AI came with the collaboration between Moorfields Eye Hospital London and Google’s AI team, DeepMind. Together, they created an AI system that can detect 50 ophthalmology diseases based on three-dimensional optical coherence tomography (OCT) data.

AI algorithms can help in the segmentation and classification of ophthalmic imaging data such as color fundus photography (CFP) for various eye diseases, such as diabetic retinopathy (DR).

Kantify’s work in early detection of Diabetic Retinopathy

Recently, Kantify has developed an Artificial Intelligence model to detect diabetic retinopathy based on patient’s imaging data. The model is showing impressive results, with both high sensitivity and specificity.

Why is it important?

Diabetic Retinopathy (DR) is the leading cause of blindness for people having diabetes. It affects the blood vessels in the retina (the light-sensitive layer of tissue in the back of your eye).

Diabetic Retinopathy may not show any symptoms at first, but early detection can help patients take steps to protect their vision. If not detected and cured on time, DR can lead to other serious eye conditions, like developing diabetic macular edema (DME), neovascular glaucoma, or retinal detachment.

With more and more people affected around the world, DR is deemed as a global public health problem. Therefore, a large and cost-effective screening for DR is needed to detect potentially threatening changes in the eye, at early stages. As we all know, early intervention is the most cost-effective choice both for patients and medical professionals.

What are the benefits of using AI for early detection?

AI-assisted screening and diagnosis of common diseases in ophthalmology can assist doctors in their work. From a patient’s perspective, AI reduces obstacles to access a point of care where an ophthalmologist is not available. Ultimately, the implications of AI in ophthalmology can improve the accessibility, availability, and productivity of eye care services.

Timely detection of pathologies

Quick and accurate medical diagnoses are crucial for the successful treatment of diseases. Using machine learning algorithms capable of analyzing a patient's image results, medical practitioners are able to improve the timely detection of ocular anomalies, leaving them more time to focus on the patient's treatment. AI for early detection is not about replacing medical professionals, but about:

• Assisting them in their work;

• Helping them confirm their diagnosis;

• Enabling them to screen more patients.

Reducing costs and increasing efficiency

While the process of creating a fundus photograph for a patient is cheap, analyzing the results is expensive - usually requiring the intervention of a trained ophthalmologist. On the other hand, the cost of running the fundus imaging results through an AI algorithm is very low. Even better: AI algorithms have no limits in terms of processing capacity, which means it can quickly process a large amount of data, enabling medical professionals to meet patients’ needs with high-speed. This increases, for example, the probability of identifying early-stage disease that may only show features in small, isolated areas.

Improving healthcare for patients and practitioners

Using AI and Digital Health solutions can enable more patients to have access to accessible care. In addition to advancing early detection of a disease, AI may contribute to a better understanding of disease mechanisms and development. Deep learning AI has the potential to identify previously unknown patterns of a disease that would improve the practitioner’s understanding of the pathogenesis of the disease, and provide additional markers for diagnosis, staging, and prognosis.

More Information

AI is transforming healthcare, and early detection of diseases is one of the most promising areas of development. By powering a new generation of systems that equip clinicians with smart tools, AI will lead the way in a new era of exciting breakthroughs in patient care. Let’s get in touch to discover how you can use AI for early detection of ocular pathologies.