menu

News & Events

  • Title : HUMC adopts AI-based facial recognition system for patient safety for the first time in Korea
No. :
476
Date :
2019.02.08
Hit :
12066

content


  • Hallym University Medical Center adopted AI-based facial recognition system for patient safety for the first time in Korea, as a medical institution.


    Since the introduction of the system in October last year, HUMC has been using this technology to identify its staff members. The use of facial recognition system will be expanded to the field of patient safety including the identification of patients in operation rooms and various examination rooms at its hospitals within this year.


    A facial recognition system uses biometrics to map facial features from a photograph, like the angles between the line drawn from the top most sunken part of the nose bridge to the tip of the nose and the line from the tip of the nose to the chin and such. This "faceprint" is saved into database, and the AI uses this data for identification and authorizations.


    The Information Management Division at HUMC developed this technology itself, which detects and compares human faces with the facial features from the image saved in the database.


    Within the system, the technology developed by NEOSECU, a company that provides biometric solutions for identification, was used that saves facial features within 0.3 seconds onto a user's profile in the database.


    Meanwhile, the use of facial recognition system will be extended to OCS and EMR at five general hospitals of HUMC within this year.


    Thanks to the implementation of AI-based facial recognition system, HUMC has become able to identify the patients with 100 percent accuracy when administering anticancer drugs and giving a blood transfusion. Officials at HUMC said that the use of facial recognition system will be limited to only the patients who agreed to use it, in order to protect personal information.


    By Chul Kwon, Int’l Cooperation Team, HUMC (chris@hallym.or.kr)


Preview

Next