Artificial Intelligence in Ophthalmology

"Artificial intelligence is around us, and it will change medicine, including ophthalmology. Come and learn about recent developments in different subfields of ophthalmology, based on AI technology!"

Andrzej Grzybowski, Professor of Ophthalmology and Chair of the Department of Ophthalmology, University of Warmia and Mazury, Olsztyn, Poland, and Head of the Institute for Research in Ophthalmology, Foundation for Ophthalmology Development, Poznań, Poland, talks about the inspiration behind the virtual event, the impressive speaker list, and his own work in the field. 

Click here to see the program and register

When did you first decide to organize this online event; what was the inspiration behind it?

I have thought about it for some time. However, the final argument for going ahead with the event was to receive the support from the Polish Ministry of Science and Education. We applied for such a grant because organizing such an international event is a demanding task. When we receive a positive response and financial support to organize it, I decided to host it this year.

We want to show why AI in ophthalmology is a very hot topic today. In recent years, it has been a very productive area of research, particularly in ophthalmology, with have hundreds of papers from different studies published every year, and many promising applications for AI in healthcare, using different approaches. This was the reason I decided we should present some of the best examples and discuss the promise of the future AI applications, but also some possible limitations.

There are some ambitious expectations for AI in healthcare and ophthalmology. And some of them include, for example, outperforming doctors, helping to diagnose what is presently undiagnosable, helping to treat what is presently untreatable. And to recognize in images what is presently unrecognizable. A good example is the 2018 study done by the group of Ryan Poplin, Google Research, Google Inc., in Mountain View, California, USA. They produced an AI algorithm that was able to – based on fundus images – recognize the patient’s age, sex, whether the patient was a smoker or not, their BMI, and systolic and diastolic blood pressure. All of these features were beyond the possibility of detection by a human grader, even very experienced retina specialists. It shows completely new prospects of this technology for healthcare, and ophthalmology in particular.

The speaker list features the best of the best in this field. How did you choose the presenters?

I am very happy that I was able to invite very well recognized great experts in the field from all over the world, to begin with Emily Chew, who is the Director of the Division of Epidemiology and Clinical Applications at the National Eye Institute, and she is a very well-known scientist from the US and Editor in Chief of the Ophthalmology Science. Then we have Michael Chiang, who is a pediatric ophthalmologist, and the Director of the National Eye Institute, very much involved with medical devices and using AI algorithms in retinopathy of prematurity. Then, again from the US, is Michael Abramoff from the University of Iowa, one of the gurus of this field, creator of the first FDA-approved autonomous AI algorithm for diabetic retinopathy. Another speaker is Linda Zangwill from University of California in San Diego who has worked for many years on a glaucoma AI algorithm, and we have collaborated on this topic.

And then we have great speakers from Asia. We have Paisan Ruamviboonsuk from Rajavithi Hospital in Bangkok,  Thailand, who has published several very important studies on the use of AI technologies with Google. Then, Daniel Ting from Singapore, one of the leaders of this subject worldwide, working not only in Singapore, but also in Guangzhou, China.

From Europe, we invited Damien Gatinel from Hospital Foundation Adolphe De Rothschild in Paris, France, whose team developed a very interesting IOL calculating formula based on AI that also shows potential for other applications, including some for retina. Then, we have a group of Polish scientists from all over the country. All of them are experts, computer scientists or vision researchers working on AI-related studies.

Are you planning for this to be a regular event?

Since we received the grant from our Ministry of Science for two years, we will be organising it also next year, and then, depending on the feedback and on the needs of our attendees, we may decide to continue. I'm pretty sure that there is a need for such a meeting, so I'll be happy to continue this idea in the future.

Could you briefly tell me about your own work in this area?

My team has been working with AI for some time, mostly in diabetic retinopathy, and we have published a few studies on that. Then we moved into the practical application and we have conducted a relatively big project in Wielkopolska region, one of Poland’s regions, with 3.6 million inhabitants. We designed diabetic retinopathy screening based on AI, and we aim to screen 40,000 of diabetic patients. This project is funded by the European Union, and it will be finished this year. It’s real-world scenario project. Some clinical studies have limitations because very often they can be hardly replicated in the real life so this is what we decided to conduct in Poland.

We also collaborate with many other groups all over the world, including Singapore and Thailand, as well as Beijing, China, with Ningli Wang, to use machine learning to determine risk factors for myopia progression. We did a few studies based on comparisons between different algorithms, which I believe is very important, but still relatively rare, with still only a few such studies available. And, quite recently, we did the interesting study, which is not published yet, on variability of grading diabetic retinopathy screening images among untrained retina specialists in Poland. Of course, I welcome anybody who might want to collaborate with us.

We have also published a book, launched at the end of last year, on AI in ophthalmology. It’s a join project: I’m the editor, but I invited the very best experts worldwide, from all continents.

To finish off, please tell me in a couple of sentences why our readers should register for the event and join it on June 3.

Artificial intelligence is around us, and it will change medicine, including ophthalmology. Come and learn about recent developments in different subfields of ophthalmology, based on AI technology!