Health Tech World hears from the co-founder and CEO of tech startup, Mediwhale which is using AI-enabled retina scans to detect diseases before it’s too late.
Sitting across from his opthalmologist, Kevin Choi received a shocking blow.
Despite only being 24 years old, he had lost 60 per cent of his vision from advanced glaucoma.
If it had been diagnosed and treated earlier, he could have saved his vision.
“I had no symptoms, I had no clue about it,” Choi told Health Tech World. “It was unexpected.”
Despite living in a country known for its excellent public health service, Choi’s story is all too common.
It is estimated that half of glaucoma cases worldwide are undiagnosed. The symptoms start slowly, developing over time.
As a result, most patients are completely unaware of their condition until it is too late.
The disease causes sight loss and often blindness in more severe cases.
“The Korean government offers a very good healthcare system and it’s easy to get those types of tests, like eye tests and screening tests,” Choi said.
“Even in Korea, [they] missed my problem in this well-structured system.”
The diagnosis was life-changing for Choi, but with it came a motivation to solve the issue so others would not have to go through the same experience.
From patient to health tech founder
Choi was a recent graduate of Pohang University of Science and Technology (POSTECH) with a major in industrial management engineering.
As he received treatment for the disease, he and his opthalmologist, Dr Tyler Rim, connected over their mutual interest in machine learning.
From this chance encounter, the seeds of Mediwhale were sown.
The pair launched the startup in 2016, driven by a mission to seek new ways to detect health risks before symptoms manifest.
Choi and Dr Rim saw an opportunity to integrate advanced machine learning algorithms with retinal fundus photographs.
“When I first met with [Dr Rim], he told me his technical vision as a retinal specialist,” Choi said.
“I had a patient perspective and we shared a vision.”
Dr Rim, with his background as a vitreoretinal surgeon, envisioned that deep learning algorithms could predict health risk factors using these retinal images.
This fueled a new direction in his research at Yonsei University College of Medicine in Seoul, focusing his attention on data sciences, biostatistics, epidemiology, big data analytics and machine learning.
With the addition of chief technology officer Young Lee, the foundational diagnostics algorithm of MediWhale evolved into a sophisticated AI diagnostics tool called Reti-Intelligence.
‘The eye is a window to the whole body’
The human eye is a remarkable organ, providing a direct view of blood vessels through the retina, the rear interior wall of the eye.
Traditionally, ophthalmologists and optometrists have used retinal photographs to diagnose various eye disorders.
However, Mediwhale recognised the untapped potential of these images in detecting more than just ocular conditions.
The technology aims to go beyond the limitations of human interpretations, employing artificial intelligence to analyse biomarkers in retinal images and predict not only eye diseases but conditions affecting other vital organs.
Reti-Intelligence leverages non-invasive retinal scans through a common fundus camera combined with trained deep learning algorithms to detect disease risks even before symptoms appear.
This initial assessment, conducted in primary care in less than one minute, can save patients time, money and the impact of diseases that otherwise could go undetected.
Quick, non-invasive and cost-effective
In validation studies, the tool has produced impressive results.
Reti-Eye shows a minimum of 96 per cent accuracy, while Reti-CKD scans yield better risk prediction outcomes than traditional blood or urine tests, the company claims.
Mediwhale’s Reti-CVD and Reti-CKD analyse specific patterns and biomarkers present in the retinal images to predict cardiovascular and kidney diseases.
“There is a huge consensus among ophthalmologists that the eye could be the window to full-body conditions related to the kidney or the heart.
“And Dr Rim understood the systemic relationship between the eye and cardiovascular changes.
“Those metabolic conditions are linked to problems on the vascular level,” Choi said.
“There are so many micro-vascular [abnormalities] that can be seen in the retina.
“Patients with metabolic disorders like hypertension and even obesity; they are the target of our tests.”
Kidney and cardiovascular disease are among the leading causes of morbidity and mortality worldwide, so giving healthcare professionals the ability to detect health risks before they arise could be transformative.
The non-invasive nature of the tool also makes it more accessible, safe and affordable compared to traditional methods such as CT scans.
“The solution is very helpful for both patients and clinicians,” Choi said.
“For people with type 2 diabetes, they need a cardiac CT scan to identify their cardiovascular risk.”
Mediwhale’s Reti-CVD is a viable alternative, he said.
The scan has been shown to predict cardiovascular risk as accurately as a cardiac CT scan while also saving costs to healthcare services and providing a more comfortable experience for patients.
“We believe that AI technology is bridging the gap,” Choi added.
“Reti-CVD [allows] a primary care physician to offer a comprehensive systemic disease screening at the frontline of care without a CT or access to a large hospital.”
Plans for expansion
Mediwhale is already in use in South Korea and approved in the EU and Australia.
As it continues to expand into additional territories, the company is continuing to gather more clinical evidence and further validate the technology.
A recent Series A funding round raised $9 million which will be used to accelerate the global commercialisation of Reti-CVD.
“We are very focused on finding more real-world data in the Korean market around patient compliance and the therapeutic effect of our tests,” Choi said.
“Our vision is helping everyone get healthier and manage their health conditions with a simple scan.
“We are striving to make this important technology available in clinical settings worldwide.”
The company is now working towards entering the US market, anticipating FDA approval by 2024.
With cardiovascular disease (CVD) as the leading cause of death for Americans, the need for better diagnostics tools is undeniable.
Presenting at the American Heart Association and joining the cohort of HeartX, MediWhale garnered significant interest, with five hospitals already agreeing to conduct clinical studies using Reti-CVD to screen patients.
Looking further ahead, Choi sees the potential of the technology reaching beyond cardiovascular and kidney disease.
And as smartphone cameras enable increasingly higher-resolution photography, the tool could be brought into people’s homes in the future, allowing individuals to track and monitor their health independently.
“The vascular system is all linked,” he said.
“So we are looking at cerebrovascular, stroke in the brain and long-term problems like COPD and liver problems such as fatty liver disease.
“These are all very important problems.”
Healthcare innovators and leaders honoured at Imprivata HealthCon user group and awards ceremony
Wearable communication system may reduce digital health divide
Molecule trains the immune system to prevent cancer
Diabetes patients urged to use fitness games with caution
Urgent work needed to tackle ‘substantial’ digital health inequality
eSight: “The technology has the potential to change someone’s life”
Microsoft invests £2.5 billion in UK AI
TMS shows promise in tackling depression ‘epidemic’
AI depression app set for NHS clinical trial
UK Biobank releases world’s largest single set of sequencing data
- News3 weeks ago
Why a leading healthcare CEO sees recombinant DNA as a metaphor for developing breakthrough technologies
- AI6 days ago
AI model predicts breast cancer risk without racial bias
- Medtech4 weeks ago
Surtex Instruments to unveil game-changing Infinex microsurgery instruments at MEDICA
- AI2 weeks ago
Humans make better cancer treatment decisions than AI, study finds