‘I want to innovate diagnostics with world’s best AI technology’

Medipixel CEO Song Kyu-seok previously worked as an IT developer at LG Electronics and AhnLab. He developed Medipixel XA, an automated cardiovascular image analysis program that uses AI deep learning technology to create a better treatment environment.

“I am a developer who has become an entrepreneur. I’ve developed products that almost every Korean household uses, such as AhnLab’s V3 vaccine (an anti-virus software program). The same applies to the health sector. I know how to make a health product that every hospital uses. “

Medipixel CEO Song Kyo-seok, having faith in his eyes, drew blood vessels on a piece of paper to explain the company’s medical diagnostic program.

In an interview with Korea Biomedical Review, he said he didn’t mean to sound presumptuous.

When asked about the company’s technology, however, he said it was “the best in the world”.

“Medipixel XA, an automated program for analyzing cardiovascular vessels, analyzes cardiovascular images in the shortest possible time with top performance. I’m sure our technology is the best in the world in the current era, ”said Song.

Medipixel XA, a software program with deep learning, takes only a second to recognize cardiovascular contours and lesions.

There is no need to manually set the division area or perform a new operation for each lesion.

The program identifies the major blood vessels and individual lesions in branch blood vessels and recommends a bespoke stent. It plays its role as an assistance device that is used in real time on site.

“When using existing products, doctors had to wait three to five minutes for results after manually setting the segmentation area. So how can you use them in urgent surgery? These products were supposed to help the operation, but they didn’t make it into the operating room, ”said Song.

After hearing of such difficulties from surgeons, Song developed Medipixel XA with the assistance of cardiologists at Asan Medical Center (AMC). Song’s 2017 AMC medical data analysis competition win helped him enlist AMC’s support.

“We were incredibly lucky,” recalled Song, saying that in addition to a lot of accumulated data, the AMC also had annotation data, which is essential for supervised learning.

According to Song, the company could perform AI deep learning by securing more than 7,000 images from well-refined videos related to cardiovascular interventions.

The lesion can be automatically recognized by AI and displayed as a quantified number per pixel. The company name “Medipixel” was derived from this.

Medipixel is also working on a navigation automation technology in the cardiovascular system in which a guide wire will automatically find the legion.

Song predicted that AI technology would be widespread in the robotic surgery market and went a step further.

Song explained that with artificial general intelligence that predicts and reacts on its own outside of the learning environment, the guidewire in any blood vessel could be steered in any desired direction.

“Current AI / deep learning technology focuses on diagnosis and prediction in the medical field. But the demand for treatment-focused AI programs will increase as the robotic surgery market expands, ”he said. “That is why our company relies on ‘legal aid’.”

How do doctors react to Medipixel XA? The program received medical device certification from the Ministry of Food and Drug Safety in January. Clinics in cooperation with Medipixel will start clinical studies in the second half of the year.

A single screen of Medipixel XA shows the cardiovascular contour and lesion site, as well as the recommended stent types.  (Image credit: Medipixel).
A single screen of Medipixel XA shows the cardiovascular contour and lesion site, as well as the recommended stent types. (Image credit: Medipixel).

“Even those who were skeptical have welcomed our prototype product after using it. The more it is introduced in clinical centers, the more enthusiastic their responses will be, ”said Song.

“The reimbursement of AI medical devices should be viewed from a future-oriented perspective”

While Medipixel is taking steps to commercialize the program, it still has one major challenge to solve – obtaining health insurance for the program.

So far, no AI medical device approved by the Ministry of Food and Drug Safety has been reimbursed.

Medipixel and other AI medical device developers share concerns that it is extremely difficult in Korea to create a profitable model through reimbursement.

“Of course, the government needs to be careful about spending the health insurance budget, which comes from taxpayers’ money,” said Song.

The U.S. and European countries where AI technology has advanced handle reimbursement issues in different ways, he said.

“What I want to suggest is that we can do the review and make it a little more forward-looking,” he said.

More specifically, Song suggested setting up a pilot for a year or two for products that are safe because they are not administered to the human body.

According to Song’s suggestion, if they benefit patients and help reduce costs in the clinical setting during the pilot, they can obtain regulatory approvals and reimbursement.

That way, medical institutions have the opportunity to take advantage of new technology and companies can collect clinical data, he said.

In Korea, however, the discussion is moving slowly. Because of this, AI medical device developers are looking for opportunities abroad.

Medipixel is negotiating with cardiovascular imaging companies or surgical device companies to provide embedded medical device software. In the highly competitive medical device market, there is always a high demand for newer software that is faster and more accurate.

“We have little room for a breakthrough in medical hardware, but a lot of potential in software. So this is a once in a lifetime opportunity to move forward in the world market, ”said Song.

AI is becoming part of life, AI developers are becoming global

Song emphasized that people should not vaguely view AI technology as the “next generation technology of the future.”

“AI is already around us. It’s so natural that we’re not even aware of it. People in all medical fields are thinking about how to apply AI and what products to develop. This is not an optimistic view as a developer, but it is the reality, ”he said.

Korea, which used to be a laggard in AI technology, has made rapid strides in just two or three years. Numerous companies boasted technological might, and Korea made a name for itself in the global healthcare market.

Thanks to the country’s advances in AI, Song has an international conference call several times a week.

Medipixel attracted 6.5 billion won ($ 5.8 million) investments on May 13. The Ministry for SMEs and Startups previously called Medipixel the “national representative of innovative companies”.

The company’s internationalization is part of Song’s pursuit of happiness. If a lot of people use the product that he made with his own hands, it will make him happy as a developer and engineer, Song said.

“I hope that more healthcare professionals will be able to use our product and that more patients will be treated with our product. If that happens, I hope Medipixel could be a company that creates a better diagnostic and treatment environment. “

Comments are closed.