close x
download

Brochure

Enter Your Info Below And We Will Send
You The Brochure To Your Inbox!

aibotics go-ditial brochure (en)

Thank you!
Your submission has been received!

Oops! Something went wrong while submitting the form

brochure
News

Detecting Heart Disease in the Blink of an AI

Asian Scientist
7.8.2020

Dr. Carolyn Lam, co-founder of medical technology startup eKo.ai, is hoping to turn the tide in the battle against heart disease by using artificial intelligence.

What comes to mind when you hear the word ‘ultrasound’? If you’re like most people, the word probably conjures up images of a pregnant mother lying on an ergonomic table, looking expectantly at a moving fetus on a black and white screen.

But ultrasounds are more than just a tool for monitoring a baby’s health during pregnancy. According to cardiologist Dr Carolyn Lam, co-founder of medical technology startup eKo.ai and professor at Duke-NUS Graduate Medical School, ultrasounds also provide valuable insights into the inner workings of the heart.

These heart-specific ultrasounds are known as echocardiograms, and are widely used by cardiologists like Lam to diagnose problems like blocked valves and stiff heart muscles. Given that cardiovascular disease is the world’s biggest killer, accounting for about 17 million deaths every year, early detection of cardiovascular risk is needed to lower this death toll. However, shortages in skilled technicians, expensive software and difficulties in data interpretation limit access to life-saving tools like the echocardiograph.

Echocardiograms—or echo for short—are the doctor’s first tool of choice to image the heart,” explained Lam. “But it’s also hard to analyze and therefore remains in the hands of specialized cardiology centers.

Interpreting an echocardiogram can be quite burdensome, with a single analysis taking at least 30 minutes and 250 clicks due to the manual sorting of images, tracing of heart borders and multiple measurements involved—all of which are error-prone processes. Even worse, data analysis was often subjective, with different doctors often giving vastly different diagnoses from the same echocardiogram.

For this reason, Lam co-founded eKo.ai in 2017 to automate the fight against heart disease with the help of artificial intelligence (AI). With eKo.ai’s platform, the entire echocardiography process—from sorting to measuring—could be done in a single click.

“I became convinced that AI can automate the highly manual error-prone way we do things, making me more efficient and accurate at what I do,” she said.

Through a process called machine learning, a computer program equipped with learning algorithms can improve its ability to find patterns in existing data and interpret new data. With its access to millions of echocardiogram images, eKo.ai’s AI algorithm is trained to recognize features and patterns in ultrasound images and make standardized predictions in roughly one minute, according to Lam.

eKo.ai’s innovative technology has already caught the attention of several big players in the pharmaceutical space and beyond. Earlier this year, the startup established a strategic partnership with AstraZeneca, one of the world’s biggest pharmaceutical companies. They’ve also raised US$4 million in funding from backers including Sequoia India, EDBI and SGInnovate.

In light of the ongoing coronavirus pandemic, eKo.ai’s work has never been more relevant. After all, COVID-19 patients with underlying heart disease have been shown to have a higher risk for infection and death. Meanwhile, COVID-19 can also induce conditions like irregular heartbeats and increased blood clots. Because of this, eKo.ai is now in the process of signing several new research agreements with their pharmaceutical collaborators, revealed Lam.

Moving forward, Lam and her team at eKo.ai envision a more democratic future where echoes could be easily accessed and analyzed even in healthcare settings with limited resources. But for this to happen, non-specialists need to be equipped with smart software like eKo.ai to accurately distinguish between normal and abnormal findings.

“We will accomplish this goal by pairing our AI algorithms with handheld ultrasound devices that that can plug into smartphones,” she explained. “The future is mobile.”

For more related stories visit Asian Scientist

...
more posts