US 12,114,983 B2
Apparatus and method for cardiac signal processing, monitoring system comprising the same
Jong Ryul Yang, Gyeongsan-si (KR); and Ju Yeon Kim, Daegu (KR)
Assigned to Industry Academic Cooperation Foundation Of Yeungnam University, Gyeonsan-Si (KR)
Appl. No. 17/256,815
Filed by INDUSTRY ACADEMIC COOPERATION FOUNDATION OF YEUNGNAM UNIVERSITY, Gyeonsan-si (KR)
PCT Filed Nov. 6, 2020, PCT No. PCT/KR2020/015443
§ 371(c)(1), (2) Date Dec. 29, 2020,
PCT Pub. No. WO2021/096162, PCT Pub. Date May 20, 2021.
Claims priority of application No. 10-2019-0144981 (KR), filed on Nov. 13, 2019; application No. 10-2019-0172336 (KR), filed on Dec. 20, 2019; and application No. 10-2020-0145736 (KR), filed on Nov. 4, 2020.
Prior Publication US 2021/0290139 A1, Sep. 23, 2021
Int. Cl. A61B 5/318 (2021.01); A61B 5/00 (2006.01); A61B 5/024 (2006.01); A61B 5/327 (2021.01); A61B 5/352 (2021.01); G06N 3/08 (2023.01); G16H 50/20 (2018.01)
CPC A61B 5/327 (2021.01) [A61B 5/02405 (2013.01); A61B 5/352 (2021.01); A61B 5/7267 (2013.01); G06N 3/08 (2013.01); G16H 50/20 (2018.01)] 10 Claims
OG exemplary drawing
 
1. Apparatus for cardiac signal processing comprising:
an input signal processor for extracting data necessary for cardiac diagnosis from a sensor signal of a radar sensor measured in a non-contact state by processing into an electrocardiogram signal that is an electric signal;
a signal processor for performing learning to detect a heart rate variability per minute using a generative adversarial network algorithm for comparing extraction data extracted by the input signal processor with a reference data;
an output signal processor for outputting, by using the heart rate variability per minute learned by the signal processor, either of:
an analysis parameter based on heart rate variability analysis, or
a signal processed into a waveform of an electrocardiogram signal; and
a diagnostic unit for diagnosing a heart condition using the analysis parameters or signals processed into the waveform of the electrocardiogram signal, wherein
the generative adversarial network is a learning model comprising:
a discriminator network; and
a generator network.