Samsung Medison unveils deep learning-based breast ultrasound imaging device

2016.04.22 10:29:06 | 2016.04.22 10:29:46

[RS80A]이미지 확대


Samsung Medison Co., a medical device subsidiary of South Korea’s tech giant Samsung Electronics Co., announced on Thursday that it has released premium ultrasound system RS80A with latest upgrades that include ‘S-Detect,’ a deep learning algorithm for radiologists in breast lesion analysis.

Deep learning is a computing process that imitates the brain with digital neural networks that gather information and react to it. It is designed to identify specific patterns from myriads of data sources.

S-Detect allows users to understand the characteristics of the displayed lesion of the breast on whether it is benign or malignant with just a few clicks. The company said it has applied deep learning technology to ultrasound imaging for the first time in breast diagnosis.

The new system utilizes big data collected from about 10,000 breast exam cases and provides the characteristics of the breast lesion with more accurate and faster results by adopting a deep learning algorithm in lesion segmentation, characteristic analysis, and assessment processes.

“We expect users will be able reduce taking unnecessary biopsies, identify malignant lesions that are most likely false benign and locate an area potentially indicative of malignancy,” said professor Han Boo Kyung, a radiologist at Samsung Medical Center.

The significantly upgraded system will be available initially in Korea, Europe and the Middle East this month. Marketing will begin in other countries depending on local regulatory approval schedules.

By Song Sung-hoon

[ⓒ Pulse by Maeil Business Newspaper &, All rights reserved]