• COVID-19
  • Biosimilars
  • Cataract Therapeutics
  • DME
  • Gene Therapy
  • Workplace
  • Ptosis
  • Optic Relief
  • Imaging
  • Geographic Atrophy
  • AMD
  • Presbyopia
  • Ocular Surface Disease
  • Practice Management
  • Pediatrics
  • Surgery
  • Therapeutics
  • Optometry
  • Retina
  • Cataract
  • Pharmacy
  • IOL
  • Dry Eye
  • Understanding Antibiotic Resistance
  • Refractive
  • Cornea
  • Glaucoma
  • OCT
  • Ocular Allergy
  • Clinical Diagnosis
  • Technology

Podcast: AI/MR in ophthalmology: The future is now

Podcast

Eyenuk Inc. founder and CEO Kaushal Solanki, PhD, went from working for the Defense Department to founding a company with a focus on artificial intelligence for ophthalmology. Find out what led him to make that jump.

Eyenuk Inc. founder and CEO Kaushal Solanki, PhD, went from working for the Defense Department to founding a company with a focus on artificial intelligence for ophthalmology. Find out what led him to make that jump.

Editor’s note: This transcript has been edited for clarity.

David Hutton:

Welcome to EyePod, a podcast series from ophthalmology times in which we engage with key opinion leaders in interviews about the latest innovations in the areas of surgery, clinical diagnosis, therapeutics, imaging, device technology, gene and cell therapy, practice management, and other cutting edge topics. I'm your host, David Hutton.

My guest today is Eyenuk founder and CEO, Kaushal Solanki, thank you so much for joining us today.

In this episode, we're going to discuss the current state of AI and ophthalmology. But first, tell us what drives someone with a background in electrical engineering, working for the Defense Department to form a company with a focus on AI for ophthalmology.

Kaushal Solanki, PhD:

Thanks, David for having me. So, you know, that's where it gets personal, right. So when I was trained as a machine learning, or artificial intelligence scientist, I got my PhD from UC Santa Barbara. I was working at a startup company with my PhD advisor, working on cutting-edge work with drones and satellite imaging and whatnot.

My expertise was processing automatically information from these images to come to very actionable conclusions for the DOD, and other projects. And one day I just made a routine visit to an optometrist who was just across the street, and he now is a friend of mine, but at the time he said, Kaushal, you might have silently developing glaucoma. I'm like, What do you mean? And he said, you have a bigger cup-to-disc ratio. You go to see an ophthalmologist as soon as possible.

I tried to make an appointment for an ophthalmologist and it took like four and a half months. I remember I had the best insurance that money could buy. But still, I had to wait. So that gave me time to look at the problem, and it turns out some of the skills that I had on automatic analysis of images using artificial intelligence, could be applied to, you know, these problems in healthcare, where, you know, if people like me have problems with access, I'm sure there would be millions globally who would have something similar.

It checked out pretty soon, when I looked at this problem and related problems, diabetic retinopathy turned out to be one. Again, it hit home where my father who had diabetes for like, almost 10 years, and never had an eye exam. And so that confirms that this is really a problem of knowledge access that can be solved by using artificial intelligence by analyzing these images at the point of care. That led to founding of the company, and fast forward to today, we are very proud to have usage on five continents globally, in 18 countries and growing. So, that's the story.

David Hutton:

What are some of the trends we're seeing in AI and ophthalmology?

Kaushal Solanki:

So, let me start with looking at AI in healthcare, and I got a very fresh look at how AI could assist healthcare, and this was when the company was founded back in 2010.

You know, no one was talking about using AI in healthcare. It's been a lot more recent, where activity has picked up. I could think of it from a very basic fundamental. Looking at that, I look at three ways that AI could help healthcare.

First is what I call a spell check AI. So, it works in the background and flags any errors for existing programs.

Second, is fully autonomous, where AI is letting a healthcare professional scale, especially specialists to places where they have not been available anywhere.

Third is what I call superpower AI, which goes above and beyond what doctors could do today. It can come up with biomarkers or metrics of predictive biomarkers that could do things that doctors cannot do today.

With that framework in mind, we applied it to ophthalmology, which is also a very great opportunity, so to speak, because ophthalmology is a field where it's actionable. So you know, doctors can provide a treatment, do a cataract surgery, do something simple and a patient walks away with restored vision.

With that in mind, unfortunately, we are still in these situations where patients are losing vision without any symptoms. and, and hence, that's where AI comes in, especially for the field of ophthalmology. If only we could detect these diseases in patients early enough, we have the treatments, we have surgical tools to restore vision. Which really doesn't exist in many other fields. A patient could go through cataract surgery or anti-VEGF treatments and their vision is restored. But we need detection tools to screen a population to find these diseases. and that's where AI comes in, enabling early disease detection, for monitoring of diseases, and again, for improving accuracy of doctors using spellcheck that I mentioned.

David Hutton:

You kind of alluded to it, but what makes AI particularly attractive option for use in ophthalmology?

Kaushal Solanki:

So, like I said, ophthalmology is a field where clinicians can do something so long as diseases are detected early. Now, when we look at early disease detection for diseases like diabetic retinopathy, anybody with diabetes, which is 500 million patients worldwide, is vulnerable to losing vision. But we do not have enough specialist resources and healthcare professionals to screen this large growing population.

That's where AI could come in a fully autonomous manner, to detect disease, identify those patients who need further monitoring or treatment. They can go see an ophthalmologist in time and get the diseases treated. If you look at the regulatory landscape, ophthalmology is the only field where fully autonomous AI is cleared by the FDA, or global regulatory bodies. That is also something that brings to fore the advancements that AI in ophthalmology has made.

David Hutton:

Whenever AI is brought up often the initial thought is that it's going to replace humans. Tell us how this technology is actually designed to help humans be more efficient?

Kaushal Solanki:

I think what I would say is that this technology is designed to help humans be a lot more effective. So, then I would go back to what I mentioned previously, AI could provide a spell check, and nobody in the world would say spell check in a word processor is not useful. For those who have an existing system in place, AI could work in the background, make them more effective.

For situations where diabetes patients are not seeing their eye care doctor, we could make it autonomous and let the screening happen at the point of care. This could be primary care clinics or it could be pharmacies, where the mass screening or mass disease detection could happen, and those specialists could come in.

We can start using superpower AI, which is, you know, monitoring of smallest changes between successive visits, and using that to predict what's going to happen in the future so that we start helping these patients a lot more.

David Hutton:

Excellent, just how can AI be used by ophthalmologists ultimately to provide better outcomes for their patients?

Kaushal Solanki:

So again, like I said, going back, AI could work like a spell check, especially for ophthalmologists. AI could also help predict outcomes, which is also something, our group is working on. Looking at the smallest of changes that happen between successive visits and quantify it, which is something that doctors really cannot do on their own. It just takes a lot of time and effort, which is not practical. But we will not just look at these changes, but we will also use them to predict the potential outcomes for patients.

It can also help ophthalmologists by offloading the routine screening to primary care, which is again, today, people like me needing to wait four and a half months, and then just finding out that I'm fine. It's not the best use of their time. Those routine screenings can happen in primary care offices. Those who need treatment, that's where ophthalmologist and retina specialist or glaucoma specialist could be a lot more effective.

David Hutton:

What are some of the eye diseases that AI can now screen for?

Kaushal Solanki:

AI can already screen for multiple vision-threatening diseases like diabetic retinopathy, where anyone with diabetes is vulnerable to vision loss that progresses without any symptoms. It can also screen for age-related macular degeneration, the leading cause of blindness in seniors. You know, typically anybody over 55 is vulnerable to this macular degeneration, and that can also be detected - progresses without any symptoms, like other diseases that can be detected.

The third is glaucoma, where it is, you know, it is also one of the leading causes of blindness globally, and especially for African Americans, and signs of this disease can be detected. Now, these diseases are already approved, and in use in global markets, but not yet in the US, only diabetic retinopathy is cleared.

The human eye is a unique organ in our body that can provide non-invasive imaging of the microvasculature, the cardiovascular system, the optic nerve head, which is an extension of the brain. So, looking into the future, this also provides detection, early detection of diseases like Alzheimer's disease, or cardiovascular risk, or chronic kidney disease, and many of these chronic conditions that manifests in the body because of issues in microvasculature.

So, if we have microvasculature issues in the eye, you can be sure that a similar mechanism is happening in the kidneys or in the heart. So, we already have early stages of work on Alzheimer's disease, for example, working with Mayo Clinic on defining structural markers that can predict or detect worsening of the brain before any actual dementia-like symptoms occur. So that's just one example of the things that can be seen in the eye, and a final comment here is that, as I said, this retinal imaging with AI has the potential to be a diagnostic platform of tomorrow.

David Hutton:

You mentioned that you have more approvals globally than we do here in the U.S. Is there any timetable for getting more approvals here in the U.S.?

Kaushal Solanki:

We are absolutely working on getting more approvals in the U.S. and that makes our system a lot more powerful. We recently announced approval of AMD and glaucoma in addition to diabetic retinopathy in the European Union. There are systems installed, people are trained to take these images, and it's just a click of a button that they start getting a lot more powerful reports.

So, Previously it was one disease, and now they can get three diseases, they can also help lot more patients. So previously, they could only screen people with diabetes, now they can also screen seniors. It is multiplying our impact in many ways and we absolutely are looking at bringing the same in the U.S. market. We are working with the FDA. It's hard to predict the timeline, but it's going to be a year or two, before they hit the market.

David Hutton:

How has it been received in Europe?

Kaushal Solanki:

We are in the process of launching it to our select customers, especially the existing customers. So it's currently a soft launch, but our customers are very excited to have these capabilities because they can screen a lot more patients. They also feel good that we are covering many more diseases for patients.

David Hutton:

And lastly, as you look to the future, what do you see the outlook for the next couple years for Eyenuk?

Kaushal Solanki:

I think the future is about growing our footprint. It is also about, like we discussed, adding more and more detection capabilities to our platform. And to that, we are converging towards what our vision is, which is to screen every eye in the world for multiple conditions.

We believe that retinal imaging paired with AI is set to be the diagnostic platform of the future. That is because in a non-invasive way, we can see a lot of things that may be happening in the body. You know, think blood tests, but non-invasive. Eyenuk is well-positioned to play an instrumental role as we get there. Specifically, in the next two years, as I said, we want to grow our footprint globally and especially in the U.S., and we want to add new disease indications in both of the things we are making solid progress on.

David Hutton:

Thanks for listening to this episode of EyePod by Ophthalmology Times. If there are topics you would like to hear about, let us know. You can also stay connected with us on Twitter, LinkedIn or Instagram. We'll see you next time.

Related Videos
© 2024 MJH Life Sciences

All rights reserved.