Diabetic Screening

On a mission to democratize eyecare, Toku Eyes takes its technology to the U.S.

On a mission to democratize eyecare, Toku Eyes takes its technology to the U.S. 768 512 Toku Eyes

The company uses AI to detect eye health, as well as diabetes and cardiovascular conditions

There are 43 million people in the world who are blind, and another 295 million who have moderate-to-severe visual impairment. The shocking number is how many of those people have a disease that was preventable or treatable: nearly three quarters, meaning there are nearly 300 million people who should have their sight, but don’t.

One of the biggest obstacles is access to care, according to Ehsan Vaghefi, co-founder and CEO of Toku Eyes. His own father went blind as a child, spurring him to develop of an artificial intelligence platform for more accurate, and more accessible, healthcare diagnostics through eye scans.

“As I grew up, and learned more, it was interesting that a lot of these diseases of blindness that I see in my friends or their family members are very much preventable. It’s just they don’t have access to care; most of the time, they don’t have access to even the first screening, the first detection,” he said in an interview.

“My research has always been about coming up with novel technologies that can enable somebody to have an eye exam, cheaply, affordably, everywhere, so you wouldn’t need a clinician, because the clinician is probably the most expensive part of the system.”

The Auckland, New Zealand-based Toku Eyes achieves this through its two products: the first one, called THEIA, is an AI platform that can look at the images of the eye, without the need of a clinician, and quickly tell if  they need to go and see a clinician. The second product is called ORAiCLE, and it uses AI to recognize minute changes in the blood vessels to capillaries to determine a person’s risk of having a cardiovascular event, such as a stroke or heart attack.

The company, which is currently live in New Zealand and India, and which has been used by one million people so far, announced on Monday that it will be expanding its services to a new market: the United States.

Detecting heart disease through the eyes

While the idea behind THEIA was to provide patients with access to a simple eye health check through their primary care doctor, the company soon realized that it could detect more than conditions like diabetic Retinopathy, AMD, and cataracts; it could actually detect cardiovascular disease through changes in the blood vessels in the back of the eye.

“These changes are tiny and the clinician’s eye is not trained to look for them in an image, but the artificial intelligence can see them. So, about a year ago, we realized that our AI is not only showing us disease of the eye, sometimes it’s showing us something a comorbidity that is related to a disease of the eye,” said Vaghefi.

“Many times a blinding condition is a surrogate marker for something else. For example, you have hypertension, and it leads to bleeding in your eye, which leads to blindness, or you have diabetes, which leads to leakages of blood into the eye that leads to blindness.”

While eye doctors and ophthalmologist only tend to see these changes at the very end stages of the disease, when the problem has already gotten to the point where it’s noticable, ORAiCLE can detect it the beginning, years before the disease manifests itself, meaning that the patient can get a head start on the potential problem before it’s too late.

Toku Eyes’ platform is already being used as part of diabetic screening service in both the public and private sectors in New Zealand. It is also live in 20 clinics in India with the goal to be in 70 by the end of the year. So far, Toku’s products have been used by around one million people across both countries.

So far, THEIA has shown to be 96% accurate in its diagnosis, as compared to 90% for general ophthalmologists, while ORAiCLE is 25% more accurate in calculating cardiovascular risk than assessment tools on the American Heart Association and Mayo Clinic websites.

In terms of ROI, Toku has been shown to reduce the cost of screening, and increase the capacity of the same clinic, by 50% in New Zealand.

Expanding to the U.S.

While THEIA currently acts as a diagnostic tool for eye diseases, ORAiCLE is classified as a risk assessment tool (though the company hopes to change that over the next few years). As such, only ORAiCLE will initially be available in the U.S., as the company is currently seeking FDA approval to bring THEIA to market as well, something that Vaghefi expects to happen by the end of this year.

In the meantime, Toku has entered into partnerships with two companies to help bring its technology to patients in America, the first being EyeCheq, a company that puts robotic kiosks in retail locations; the company will be incorporating Toku’s software so that patients will be able to walk up to a kiosk and immediately capture an image of their eye.

“They wanted to have a whole team of clinicians in the background, assessing all the images that are coming through from these kiosks. Our collaboration with them is that you don’t need a team of clinicians, you just need one or two auditors, and the AI will just go through all the images and alert those that need to be seen by a clinician,” Vaghefi explained.

“Also, we can also identify people that are at high risk of cardiovascular events, and then work with that retailer, so the location that the camera is in, to see what other services can be provided for that person, including consultation, supplementation, and education.”

The other partnership is with Unified-Imaging, an image repository platform that gathers images from camera manufacturers. Toku will work with the company to be more of a data analytics platform.

“Based on these images, you can tell so much about this person. So, not only that person might get a referral, they might also get to be connected to these other points, like cardiology or nephrology,” said Vaghefi.

Through these partnerships and more, Toku eyes is planning, subject to finalizing regulatory clearances, to be present at more than 1,500 locations across the U.S. by 2025.

Of course, there are challenges that come with entering the U.S. health market, not the least of which is determining who will pays for this. Unlike New Zealand, which has a public health system, meaning the government pays for everything, the U.S. health market is much more complex.

“My day and night is consumed with who pays. The United States is interesting because there is a CPT code for use of AI for analyzing eye images. That’s one path; that code has been adopted by Medicare and Medicaid. So, we are looking into that but, at the same time, we believe that the cardiovascular product is very much an insurance or value-care provider product,” Vaghefi explained.

“It can tell an HMO or an insurance company, ‘the overall risk of your population is this and this is the 5% of your population that’s going to consume 50% of all the cost. If you focus on this population, and reduce the incidence rate in this 5% of the population, you’re going to save more than 50% of your entire intervention budget.’ We think there is an angle there. But, again, there’s a bit of work to be done.”

Making eye care accessible to everyone

All of this will help Toku’s larger mission, which is to expand access to care, and to help prevent treatable diseases from turning into something that causes blindness, or other serious health problems.

It’s for that reason that Toku has purposely made it so that its software can work with any camera, so that anyone would be able to access it and use it.

“Some other AIs are very much limited to high level, good quality, best grade cameras. We don’t do that. The technology is only good if it’s accessible to the lowest socioeconomic group in the society, i.e. the person who may not be able to afford to go and see a specialist. They may not even afford to go to see a doctor, but they should have access to technology,” Vaghefi said.

“If it’s not accessible in your Walmart or CVS or something like that, then it’s not meeting my standards. So, that’s why we put a lot of time and effort into our AI that can work with a lower grade, probably lower technology, camera that a CVS can afford to put in in their locations, not the $50,000 camera used by your eye specialist, but the $5,000 camera that can sit in your subway station.”

In fact, the company is eventually planning on allowing patients to capture images themselves on their own mobile phones, something that will likely be more widely adopted due to the rise in virtual care and telemedicine during COVID.

This should finally make it so that more people don’t run into the same situation as Vaghefi’s father and lose their sight to a disease that could have been prevented if it was caught early.

“What bothers me, and what keeps me going, is that my father lost his eyesight 60 years ago and, today, the same thing can happen to another person, another child that just doesn’t seem to have the access. We have had 60 years of technology innovation and nothing changed in healthcare. That’s the bit that bothers me,” he said.

“It’s all about making sure that somebody who’s coming from a low income family, maybe in a remote place, not a very well educated family, they still have access to the basic level of health, being an eye check or a cardiovascular check. Nobody should die, or go blind, because they didn’t have access to a simple check. That’s just not acceptable anymore.”

Article originally published on www.vator.tv

By Steven Loeb

NZ-based AI firm launches tool to assess heart risk through retinal scans in U.S.

NZ-based AI firm launches tool to assess heart risk through retinal scans in U.S. 768 512 Toku Eyes

Toku Eyes, a New Zealand-based healthcare AI company, is launching its tool that assesses heart risk through a retinal scan in the U.S.

The tool, called CLAiR, uses an AI platform to identify cardiovascular threats more accurately than existing risk calculators, the company claims. The platform recognizes subtle changes in aspects like blood vessels and pigmentation to identify a person’s risk of a stroke or heart attack in the next five years.

Many Americans with diabetes are at risk of long-term complications like heart disease and blindness. Early diagnosis is critical for mitigating morbidity, but diagnosis has historically been a challenge.

“If you can see the future, you can change it,” Ehsan Vaghefi, co-founder and CEO of Toku Eyes, said in an announcement. “By looking inside the eyes, we get an in-depth view of what is happening inside the entire body to better assess the risk factors of each individual and identify high-risk individuals before their condition worsens. Our goal is to make health screening simple and easy to access for the entire population so we can get in front of underlying health risks and improve patient outcomes.”

Since using a retinal camera and the AI software requires minimal training, it is more cost-effective and accessible, the company argues. The AI platform uses an image of the back of the eye to do its assessment. It then provides personalized health guidance or recommends a specialist referral. The company says the technology also works accurately with low-resolution images.

Toku Eyes is also aiming to launch a separate tool to detect blinding conditions in the U.S. by the end of this year, pending FDA approval. As part of its expansion to the U.S., the company is partnering with EyeCheq, a company Toku Eyes says is building out a network of self-service retinal image kiosks across the U.S. Another partner is Unified-Imaging, a cloud-based solution for capturing image data.

The company’s platform is currently used as part of diabetic screening services in the private and public sectors in New Zealand, according to its press release.

CLAiR is being released as a wellness device and therefore does not require FDA clearance, the company told Fierce Healthcare. This classification will also enable Toku Eyes to share cardiovascular risk data with payers to help improve the overall health of their members. Altogether, Toku Eyes hopes to be at more than 1,500 locations across the U.S. by 2025.


Article originally published on www.fiercehealthcare.com

By Anastassia Gliadkovskaya

New Zealand tech company develops tool to detect diseases like diabetes

New Zealand tech company develops tool to detect diseases like diabetes 768 512 Toku Eyes

Around 280,000 New Zealanders have diabetes, with the disease almost three times more common in Māori than non-Māori. And mortality rates for the disease are nine times higher for Māori aged 43-64 years than non-Māori.

Health experts say early screening is essential, but the problem is that just 35 percent of Māori opt-in for these checkups.

But a New Zealand tech company is hoping to help curb the problem as well as other diseases using its artificial intelligence platform THEIA.

Feast your eyes on THEIA, the latest eye imaging diagnostic device.

It’s a way to detect eye diseases and other diseases like diabetes.

With a retinal photo, the software uses hundreds of thousands of data points to give risk predictions, whereas currently to get that sort of diagnosis you need to go through multiple tests and doctor visits.

“I truly believe we can change eye care first and general health care second,” THEIA founder and CEO Associate Professor Ehsan Vaghefi said.

That’s because the eye is the only external part of the body where we can photograph the blood vessels that feed vital organs. Combine this with THEIA and it’s a game-changer.

“We can look at the risk of diabetes in the eye and high cholesterol in the eye, and suddenly an image of your eye can tell us a lot about your general health and not just the eye,” Dr Vaghefi said.

Auckland ophthalmologist Dr David Squirell has been part of Aotearoa’s diabetic screening for over 15 years.

He said the technology is about as accurate as he can be and better than most junior doctors.

He said the artificial intelligence helps to identify those needing help urgently to adjust their lifestyle, diet or medication.

“Technology follows the patient, we want to make it easier for people to be screened because there are multiple barriers and one of them is just physical,” Dr Squirell said.

THEIA has the potential to expand existing screening capacity by around 50 percent and reduce labour costs by 30 percent

But for this Ngāi Tai descendant, it’s about better accessibility for Māori.

Diabetes is almost three times more common in Māori than in non-Māori.

“It’s completely portable and it can be run in any setting, it can be run in iwi community centres and the AI really gives us this ability to give the result without the need of a specialist,” chief commercial officer Francesca Logan said.

“We are having constructive conversations with the Ministry of Health and the Māori Health Authority and we are hoping that by next time that our technology will be accessible in many places across New Zealand,” Dr Vaghefi said.

Rather than a window to the soul, the eye is now the window to the future of your health.

NEWS – Newshub – Te Rina Kowhai
Article originally on www.newshub.co.nz

A multi-center prospective evaluation of THEIA

A multi-center prospective evaluation of THEIA 1024 683 Toku Eyes

A multi-center prospective evaluation of THEIA to detect diabetic retinopathy (DR) and diabetic macular edema (DME) in the New Zealand screening program

Purpose: to assess the efficacy of THEIA, an artificial intelligence for screening diabetic retinopathy in a multi-center prospective study. To validate the potential application of THEIA as clinical decision making assistant in a national screening program.

Methods: 902 patients were recruited from either an urban large eye hospital, or a semi-rural optometrist led screening provider, as they were attending their appointment as part of New Zealand Diabetic Screening programme.

These clinics used a variety of retinal cameras and a range of operators. The de-identified images were then graded independently by three senior retinal specialists, and final results were aggregated using New Zealand grading scheme, which is then converted to referable\non-referable and Healthy\mild\more than mild\vision threatening categories. Results: compared to ground truth, THEIA achieved 100% sensitivity and [95.35%-97.44%] specificity, and negative predictive value of 100%. THEIA also did not miss any patients with more than mild or vision threatening disease. The level of agreement between the clinicians and the aggregated results was (k value: 0.9881, 0.9557, and 0.9175), and the level of agreement between THEIA and the aggregated labels was (k value: 0.9515).

Conclusion: Our multi-centre prospective trial showed that THEIA does not miss referable disease when screening for diabetic retinopathy and maculopathy. It also has a very high level of granularity in reporting the disease level. Since THEIA is being tested on a variety of cameras, operating in a range of clinics (rural\urban, ophthalmologist-led\optometrist-led), we believe that it will be a suitable addition to a public diabetic screening program.

Ehsan Vaghefi, Song Yang, Li Xie, David Han, David Squirrell

Download Paper from Cornell University here

You're viewing our
Europe site.

For other regions, please click below: