US 12,343,177 B2
Video based detection of pulse waveform
Jeremy Speth, South Bend, IN (US); Patrick Flynn, South Bend, IN (US); Adam Czajka, South Bend, IN (US); Kevin Bowyer, South Bend, IN (US); Nathan Carpenter, Washington, DC (US); and Leandro Olie, Washington, DC (US)
Assigned to Securiport LLC, Reston, VA (US); and University of Notre Dame du Lac, South Bend, IN (US)
Filed by Securiport LLC, Washington, DC (US); and Department of Computer Science and Engineering University of Notre Dame, Notre Dame, IN (US)
Filed on Feb. 3, 2022, as Appl. No. 17/591,929.
Claims priority of provisional application 63/145,140, filed on Feb. 3, 2021.
Prior Publication US 2022/0240865 A1, Aug. 4, 2022
Int. Cl. A61B 5/00 (2006.01); A61B 5/024 (2006.01); A61B 5/11 (2006.01); G06F 17/14 (2006.01); G06V 10/25 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01); G06V 20/40 (2022.01); G06V 40/16 (2022.01)
CPC A61B 5/7278 (2013.01) [A61B 5/02405 (2013.01); A61B 5/1128 (2013.01); A61B 5/7235 (2013.01); G06F 17/141 (2013.01); G06V 10/25 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01); G06V 20/46 (2022.01); G06V 40/161 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method for generating a pulse waveform, the computer-implemented method comprising:
capturing a video stream including a sequence of frames;
processing each frame of the video stream to spatially locate a region of interest;
cropping each frame of the video stream to encapsulate the region of interest;
processing the sequence of frames, by a 3-dimensional convolutional neural network, to determine the spatial and temporal dimensions of each frame of the sequence of frames and to produce a pulse waveform point for each frame of the sequence of frames;
modifying the temporal dimension of at least one frame with one or more dilations; and
generating a time series of pulse waveform points to generate the pulse waveform of a subject for the sequence of frames.