SKT expands utilization of machine learning to genomic analysis

By Song Su-hyun
  • Published : Aug 20, 2018 - 16:58
  • Updated : Aug 20, 2018 - 16:58

SKT's IoT/Data Unit Head Chang Hong-seong (left) holds an agreement with Macrogen CEO Yang Kap-seok in Seoul on Monday. (SK Telecom)
South Korea’s leading mobile carrier SK Telecom is moving to use its machine-learning technologies to analyze human genomes in order to raise the predictability of rare diseases, the company said Monday.

The telecom giant signed a partnership with Macrogen, a local precision medicine business, to develop AI solutions that would predict probabilities of critical diseases including cancers by analyzing personal genomes, and suggest treatments. 

To analyze human genomes of around 3 billion pairs, tools that are capable of analyzing at least 50 gigabytes of data to 900 GB are needed.

Under the partnership, SKT will tap into genomic data and medical information accumulated by Macrogen and establish an AI-based system that will analyze, store and manage the data.

The company’s machine-learning capabilities will shorten the data analysis time to a tenth, it said.

The two firms’ partnership will lead to commercial solutions for hospitals and other institutions to help them examine patients’ medical history, health habits and living environments.

“SKT will help reduce the cost and time to analyze genomes by using ICT technologies,” said Chang Hong-seong, head of the internet of things/data business at SKT. “This is how AI will play a significant role in leading change and innovation in the medical field.”

By Song Su-hyun (