<Anchor> In

Korea, lung cancer has the highest mortality rate compared to other cancers. This is because early diagnosis is difficult, but a study has been published that AI technology can detect lung cancer early.

This is Cho Dong-chan, a medical journalist.

<Reporter>

80-year-old Cho Doo-hwan had lung cancer surgery a month ago.

I'm recovering so fast that I don't need chemo or radiation therapy, thanks to my early diagnosis.

[Cho Doo-Hwan/Lung Cancer Surgery Patient: Since someone who knows goes to the hospital, he has already heard that he is in the 4th phase, so he is probably unable to perform surgery. After hearing that, I would have to do a test (I thought.)]

However, about 60% of lung cancer patients were diagnosed at a time when they could not be operated on, and in this case, the 5-year survival rate was only 6%.

People who have smoked for more than 30 years are recommended to do a low-dose lung CT test each year, but there are limitations.

[Kim, Hyun - Ku / Guro Hospital Professor: When Unconditional only be diagnosed with CT that costs money, and there is also the problem of radiation exposure yet to receive such benefits only a limited people]

and local researchers challenge the lung cancer diagnosis There is.

What the researchers have noticed is a small particle called this exosome, which emits light, which reflects the signal, depending on which cell is released.

When the blood grains of 20 ordinary people and 43 patients with lung cancer were fired up to 100 times per particle, the reflected signal was collected and analyzed by artificial intelligence, and it was found that the accuracy of the lung cancer can be diagnosed at an early stage of approximately 85%. .

[Choeyeonho / Korea University College of Health Science Research Vice President: exo (we used) a little (cell discharge), have much closer whether deep learning techniques try to analyze in normal cells or cancer cells by analyzing through the signal pattern]

team size We plan to establish a more solid basis through additional clinical trials that expand the number to 400.

(Video coverage: Han Il-sang, Video editing: Hwang Ji-young)