default_top_notch
default_news_top
default_news_ad1
default_nd_ad1

Lunit's AI-based software assisting breast cancer diagnosis wins regulatory nod

Lee Han-soo  Published 2019.07.31  14:31  Updated 2019.07.31 17:06

공유
default_news_ad2

Lunit said that it has received approval for "Lunit INSIGHT for Mammography (MMG)," an artificial intelligence-based software that assists the diagnosis of breast cancer, from the Ministry of Food and Drug Safety.

Lunit's INSIGHT for Mammography

The device, developed jointly with Severance Hospital, Asan Medical Center, and Samsung Medical Center, is a mammography imaging assistance software based on AI technology, which helps the doctor to diagnose breast cancer quickly and accurately by marking suspicious areas.

According to Lunit, a breast cancer diagnosis is more difficult for Korean and Asian women as they tend to have dense breasts. However, the company's device can detect such malignant tumors, as it uses a deep-learning model to study 200,000 mammograms, including 50,000 mammograms for breast cancer patients.

"Mammograms are particularly tricky to read," Lunit CEO Suh Beom-seok said. "Breast cancer is one of the most commonly diagnosed cancers in the world and accounts for 24 percent of all female cancers."

However, among the patients judged to have a malignant tumor through mammography, only 29 percent are found to have breast cancer, Suh added.

Suh stressed that the company introduced the device to enhance the effectiveness of breast cancer screening and to improve the reality of unnecessary testing through artificial intelligence.

"By using the Lunit INSIGHT MMG, radiologists will be able to increase their accuracy and reduce the re-examination rate," Suh said, adding that the company has set up a link so that hospitals can test the software free of charge.

“Anyone can test the performance of the Lunit INSIGHT MMG online," he said.

corea022@docdocdoc.co.kr

<© KBR , All rights reserved.>
default_news_ad5
default_news_bottom
default_nd_ad4
default_bottom
#top
default_bottom_notch