US 12,268,529 B2
Systems and methods for touchless detection of heart rate variability
Yin-Yu Chen, New Taipei (TW); Kai Jen Cheng, New Taipei (TW); and Yao-Tsung Chang, New Taipei (TW)
Assigned to WISTRON CORPORATION, New Taipei (TW)
Filed by WISTRON CORPORATION, New Taipei (TW)
Filed on Nov. 8, 2022, as Appl. No. 18/053,560.
Prior Publication US 2024/0156413 A1, May 16, 2024
Int. Cl. A61B 5/00 (2006.01); A61B 5/024 (2006.01); A61B 5/0245 (2006.01); A61B 5/05 (2021.01); A61B 5/11 (2006.01)
CPC A61B 5/7267 (2013.01) [A61B 5/02405 (2013.01); A61B 5/05 (2013.01); A61B 5/1102 (2013.01); A61B 5/1128 (2013.01); A61B 5/0245 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus for detecting heartbeat, the apparatus comprising:
a first and second sensor configured to simultaneously detect displacements of an object to generate a first and second signal, respectively, corresponding to the displacements, the first sensor detecting the displacements without contacting the object, the second sensor detecting the displacements by contacting the object;
a demodulator configured to receive and convert the first signal into corresponding in-phase (I) and quadrature (Q) components;
a first data processing unit configured to:
extract phase and amplitude data from the I and Q components,
extract phase difference and amplitude difference data from the extracted phase and amplitude data,
generate a first feature dataset from the phase, amplitude, phase difference and amplitude difference data;
a second data processing unit configured to:
detect peak times in the second signal, and
generate labeling data from the detected peak times; and
a first neural network configured to receive the first feature dataset and the labeling data for training a model,
wherein the trained model is used to infer inter beat intervals from a second feature dataset extracted from the first signal.