Skip to content
  • KOSPI 3006.16 -1.17 -0.04%
  • KOSDAQ 995.07 +1.37 +0.14%
  • KOSPI200 393.34 +0.16 +0.04%
  • USD/KRW 1175.8 0.20 0.02%
  • JPY100/KRW 1,031.72 3.87 0.38%
  • EUR/KRW 1,366.75 -3.12 -0.23%
  • CNH/KRW 184.04 0.22 0.12%
View Market Snapshot

Startups

Portrai to revolutionize biopsy quality through AI technology

The bio startup wants to improve biopsy quality by applying AI to spatial transcriptomics

By Oct 06, 2021 (Gmt+09:00)

Portrai CEO Lee Dae-seung.
Portrai CEO Lee Dae-seung.

Biopsy is one of the most used methods to examine a person’s illness, taking a tissue sample of a patient for analysis. Usually, the samples are thinly sliced and dipped into stains or dyes, but several parts of the tissue may get mixed during the process, which could reduce the accuracy of the results. And a South Korea-based artificial intelligence bio startup Portrai has developed its artificial intelligence technology to overcome such hurdles.

“We’ve applied AI technology to spatial transcriptomics, thereby increasing the level of analysis for cell tissues,” said Lee Dae-seung, the chief executive of Portrai, in an interview with The Korea Economic Daily on Oct. 5.

Spatial transcriptomics is a fresh method that is expected to significantly improve the biopsy procedure as it locates a cell's location within the tissue. It has also been chosen as the method of the year by Nature Methods, a sister journal to the global science journal, Nature.

According to Lee, Portrai uses AI technology to analyze the results of spatial transcriptomics. The company's AI technology boasts a high level of accuracy given that it has acquired a vast amount of data on spatial transcriptomics. So far, Potrai is the only company in Korea to offer such technology.

Since its inception in July, Portrai has drawn attention from the domestic medical industry as well as the startup industry for being an AI startup founded by four medical school graduates. The company’s CEO Lee is an eye specialist who has also developed machine learning-related products.

The company’s Chief Technology Officer is Choi Hong-yun, a professor in the Department of Nuclear Medicine at the Seoul National University Hospital. Also, the company's medical advisors are Na Kwon-joong, a professor in the Department of Thoracic and Cardiovascular Surgery at the Seoul National University Hospital and Im Hyung-jun, a professor at the Seoul National University Graduate School of Convergence Science and Technology.

Portrai initially began as a project between CTO Choi and Professor Na who identified spatial transcriptomics as a novel method for genomic analysis. The two betted on the strong growth potential of spatial transcriptomics in the medical market and reached out to CEO Lee, who had some startup industry experience.

Portrai’s technology was well-received from early on as the company received requests to aid in spatial transcriptomics research from the world’s largest genome research institute UK-based Sanger Institute. Also, the company’s potential profitability and competitive technology have attracted 800 million won ($670,000) in seed money from investors including Saehan Venture Capital.

According to CEO Lee, it’s challenging to accurately identify how drugs travel throughout the body and the effects that they cause, but the company believes it can significantly enhance the level of analysis on a drug’s performance by using AI-based spatial transcriptomics technology.

For example, the technology is effective in determining biomarkers, which have been recognized as a diagnosis method for various incurable diseases.

This year, Portrai is looking to collaborate with other bio companies. Last month, the company inked a partnership with a domestic bio company Therabest, which plans to use Portrai’s technology to develop immune cell treatments.

“Our goal is to increase the success rate for clinical trials by applying Portrai’s technology in the process of developing new drugs,” said Lee.

Write to Joo-wan Kim at kjwan@hankyung.com

Danbee Lee edited this article.

Comment 0

0/300