公开(公告)号：US20240108235A1

公开(公告)日：2024-04-04

申请号：US18274942

申请日：2022-02-04

Applicant: INDUSTRY ACADEMIC COOPERATION FOUNDATION KEIMYUNG UNIVERSITY

Inventor： Jong-Ha Lee ,  Beomjoon Kim ,  DAE JIN JANG

IPC: A61B5/026 ,  A61B5/00 ,  A61B5/021 ,  G06V10/25 ,  G06V10/60 ,  G16H30/20 ,  G16H30/40 ,  G16H50/30 ,  G16H50/70

CPC classification number: A61B5/0261 ,  A61B5/02108 ,  A61B5/7246 ,  A61B5/7275 ,  G06V10/25 ,  G06V10/60 ,  G16H30/20 ,  G16H30/40 ,  G16H50/30 ,  G16H50/70

Abstract: Proposed are a non-contact image-based blood pressure measurement method and system based on advanced visual intelligence, wherein a blood pressure measurement model is created by performing neural network training with an image sequence obtained by a webcam in a non-contact manner, and the blood pressure of a measurement subject is estimated with the blood pressure measurement model. Accordingly, since the webcam for obtaining a video, that is, an image sequence, in a non-contact manner is used, blood pressure is continuously estimated during the recording with the webcam without contact with a device or another person, and blood pressure is estimated just by being in front of the webcam, so the system is easy to use, inexpensive, and secures sufficient precision by using a neural network algorithm.