US 11,989,652 B2
Methods and systems to configure and use neural networks in characterizing physiological systems
Ali Khosousi, Toronto (CA); Timothy William Fawcett Burton, Toronto (CA); Horace R. Gillins, Toronto (CA); Shyamlal Ramchandani, Kingston (CA); William Sanders, Bethesda, MD (US); and Ian Shadforth, Morrisville, NC (US)
Assigned to Analytics For Life Inc., Toronto (CA)
Filed by Analytics For Life Inc., Toronto (CA)
Filed on Feb. 27, 2023, as Appl. No. 18/114,753.
Application 18/114,753 is a continuation of application No. 16/725,430, filed on Dec. 23, 2019, granted, now 11,589,829.
Claims priority of provisional application 62/907,141, filed on Sep. 27, 2019.
Claims priority of provisional application 62/784,925, filed on Dec. 26, 2018.
Prior Publication US 2023/0289595 A1, Sep. 14, 2023
Int. Cl. G06V 10/00 (2022.01); A61B 5/00 (2006.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); G06V 10/776 (2022.01)
CPC G06N 3/08 (2013.01) [A61B 5/7275 (2013.01); G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06N 20/00 (2019.01); G06V 10/776 (2022.01)] 14 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a processor, a biophysical signal data set of a subject acquired from two or more channels of one or more sensors, including a first channel and a second channel;
pre-processing the biophysical signal data set to generate one or more pre-processed data sets, wherein the pre-processed data set of the first channel includes a plurality of singularly isolated complete phase synchronized cardiac cycles, each isolated complete phase synchronized cardiac cycle having a signal peak and a same time window correspondence at boundaries of an isolated complete phase synchronized cardiac cycle to another singularly isolated complete phase synchronized cardiac cycle of the second channel; and
determining, by the processor, a value indicative of presence or non-presence of heart failure by directly inputting the pre-processed data set to one or more deep neural networks trained with a set of training biophysical signal data acquired from patients diagnosed with the heart failure and labeled with the presence or non-presence of the heart failure,
wherein an output data set is outputted via a report and/or a display based on the determined value indicative of the presence or non-presence of heart failure.