In line with its mission, the Editorial Board of MedTvoiLokony makes every effort to provide reliable medical content supported by the latest scientific knowledge. The additional flag “Checked Content” indicates that the article has been reviewed by or written directly by a physician. This two-step verification: a medical journalist and a doctor allows us to provide the highest quality content in line with current medical knowledge.
Our commitment in this area has been appreciated, among others, by by the Association of Journalists for Health, which awarded the Editorial Board of MedTvoiLokony with the honorary title of the Great Educator.
It is an innovative solution that saves your eyesight. The continuation of eye research for diabetics with the use of artificial intelligence has started in Poznań. Zuzanna Opolska talks with prof. Andrzej Grzybowski, president of the Ophthalmology 21 Foundation, who supervised the project.
- One of the complications of diabetes is damage to the small blood vessels in the retina of the eye called diabetic retinopathy
- The consequence of retinopathy is progressive, irreversible loss of vision
- Diabetic changes in the eyeballs can only be detected by ophthalmologic examination. Unfortunately, roughly half of diabetic patients report to an ophthalmologist too late for the treatment to be effective
Professor, the continuation of the project “Screening for diabetic retinopathy” is starting. Why was the Ophthalmology 21 Foundation interested in patients with diabetes?
First of all, because diabetes is one of the greatest medical problems of our time. It is estimated that 422 million people worldwide have diabetes, 179 million of whom are people who have not yet been diagnosed with the disease. In Poland, we are talking about about 3 million patients, the number of which, similarly to other developed countries, will increase. One of the complications of diabetes is diabetic retinopathy, i.e. damage to the blood vessels in the eye’s retina. This disease occurs in patients with type 1 diabetes and in about 60% of type 2 diabetics. Unfortunately, in Poland we do not have precise data on how many people lose their eyesight due to diabetic retinopathy. We do know, however, that diabetic retinopathy in Western Europe is the first cause of blindness in working-age people, i.e. 40+.
Are there medical treatments for diabetic retinopathy?
Yes, we are able to help 95% of patients. In Poland, laser therapy is the standard method of treatment. An alternative is anti-VEGF therapy in the form of intravitreal injections.
If the treatment is successful, why do we lose our eyesight?
The first problem is the lack of early access to the patient. In the initial stage of development, diabetic retinopathy remains completely asymptomatic. It does not cause any deterioration of vision, let alone pain. That is why it is so important to undergo periodic ophthalmological examinations, which are recommended for people with diabetes once a year. Unfortunately, the number of patients is so large that we are not able to test them all. The more that there are more and more diabetic patients, and not ophthalmologists. The second problem is the huge ineffectiveness of such a procedure – in the case of diabetes, only 1% of patients have changes at the fundus, and among them no more than 30% require treatment and we need to find them.
But how to select them?
Ophthalmologists have been wondering about this for a long time. Only two countries in the world: Great Britain and Singapore have national screening programs for diabetic retinopathy. For 15 years in the UK, every diabetic patient has had a fundus photo taken, which is then sent for evaluation by a specialist via telemedicine. The task of the so-called “Readers”, ie people assessing fundus images at a distance, but not being ophthalmologists, is the selection of patients who must go to an ophthalmologist. The DES (Diabetic Eye Screening) system is very effective, but very expensive – and therefore impossible to implement in many countries. The program costs around £ 100 million annually, and we’re only talking about screening.
Is there an alternative to distinguish between patients with lesions requiring ophthalmologic control and those without lesions?
Yes, in recent years AI (artificial intelligence) based computer programs have emerged to replace “British readers”. Our Okulistyka 21 Foundation has decided to use one of them as part of an innovative on a Polish scale, pilot project “Screening for Diabetic Retinopathy”. Last year, in cooperation with the Poznań City Hall, we organized a free ophthalmological screening test, which included 615 diabetic patients.
How did the research work in practice?
Each patient had a fundus image taken with a special apparatus (fundus camera) by trained nursing staff during a follow-up visit at a diabetology office. The entire study lasted only a few minutes, and patients received an on-site printout along with recommendations for follow-up visits to their ophthalmologist. The use of modern software not only shortened the examination time from 15 minutes to 5 minutes, but also did not require the presence of an ophthalmologist. Out of 615 people, only 16% of patients were diagnosed with diabetic retinopathy, the remaining 84% of patients received appropriate recommendations for further eye checks in the future. This shows how much AI helps to relieve the work of ophthalmologists.
What difficulties did the Foundation encounter during the implementation of the “Diabetic Retinopathy Screening”?
The basic problem was to convince doctors and nurses in diabetes clinics of the importance of an innovative solution. The second problem was “fear of new technology.” It will take 25 minutes for a 10-year-old to learn how to use a camera that takes pictures. However, for a middle-aged person who has never dealt with the use of computer devices, this is a challenge. It happened that the nurses refused us. Unfortunately, we do not have enough resources to organize the training. At the level of one clinic, we are able to cope, but extending the project to other centers in practice could turn out to be very difficult. Paradoxically, I see human and not technological limitations.
At the moment, the project is being continued, can screening be extended to the whole country in the future?
Yes, in the period from February to June 2019, another 400 patients will be included in the study, but this is not the end. We are preparing to organize a project at the level of the entire Greater Poland Voivodeship, which would cover a population of 40 people. If we succeed, it will be the largest diabetic retinopathy screening project in the whole of Europe. The next step will be to convince the Ministry of Health of the need to introduce screening for diabetic retinopathy throughout the country. Of course, I approach the Foundation’s plans with humility – it is one thing to plan a project on paper, and another to put it into practice. We are currently looking for volunteers and potential collaborators who would like to learn something interesting, but also work with us on the continuation of this project. We also invite scientists interested in this subject to cooperate. We have already prepared the first scientific publication of our research in 2018, which is already accepted for publication in the prestigious English-language journal Acta Ophthalmologica. We cooperate with the Ophthalmology Clinic in Singapore, the University of Iowa in the USA and the Ophthalmology Clinic in Reykjavik, Iceland. I invite students of optometry, medicine and nursing, as well as nurses and optometrists to cooperate. Please contact us on the website of the Ophthalmology 21 Foundation.