US 11,877,861 B2
Methods and systems for labeling sleep states
Conor Joseph Heneghan, Campbell, CA (US); Jacob Anthony Arnold, Fremont, CA (US); Zachary Todd Beattie, Pleasant Hill, CA (US); Alexandros A. Pantelopoulos, Berkeley, CA (US); Allison Maya Russell, Berkeley, CA (US); Philip Foeckler, Richmond, CA (US); Adrienne M. Tucker, Fremont, CA (US); Delisa Lopez, San Francisco, CA (US); Belen Lafon, San Francisco, CA (US); and Atiyeh Ghoreyshi, San Francisco, CA (US)
Assigned to FITBIT, INC., San Francisco, CA (US)
Filed by Fitbit, Inc., San Francisco, CA (US)
Filed on Mar. 9, 2022, as Appl. No. 17/690,369.
Application 17/690,369 is a continuation of application No. 17/560,639, filed on Dec. 23, 2021, abandoned.
Application 17/560,639 is a continuation of application No. 15/438,643, filed on Feb. 21, 2017, granted, now 11,207,021, issued on Dec. 28, 2021.
Claims priority of provisional application 62/384,188, filed on Sep. 6, 2016.
Prior Publication US 2022/0265208 A1, Aug. 25, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. A61B 5/00 (2006.01); A61B 5/024 (2006.01); A61B 5/11 (2006.01)
CPC A61B 5/4812 (2013.01) [A61B 5/02416 (2013.01); A61B 5/02438 (2013.01); A61B 5/11 (2013.01); A61B 5/1114 (2013.01); A61B 5/681 (2013.01); A61B 5/7264 (2013.01); A61B 5/7267 (2013.01); A61B 5/743 (2013.01); A61B 2562/0219 (2013.01)] 19 Claims
OG exemplary drawing
1. A computer-implemented method for labeling stages of sleep, the computer-implemented method comprising:
executing instructions stored within a non-transitory, machine-readable storage medium that is operatively coupled to one or more processors to cause the one or more processors to perform the following operations:
obtaining, using a wrist-worn device, motion data for a user during a time window in which the user is asleep;
obtaining, using the wrist-worn device, cardiopulmonary data for the user during the time window;
processing the motion data and the cardiopulmonary data with a classifier model to generate one or more labels,
each of the one or more labels generated for a corresponding time period within the time window,
each label indicating the user is awake or in one of a plurality of different stages of sleep during the corresponding time period,
extracting a first feature from the motion data for the user;
extracting a second feature from the cardiopulmonary data or from inter-beat intervals for the user,
wherein the second feature comprises at least one of a variability of a de-trended respiration rate, an inter-percentile spread of a heart rate, and a normalized de-trended heart rate,
and wherein the processing of the motion data and the cardiopulmonary data with the classifier model includes processing the first feature and the second feature to generate the one or more labels;
and displaying, on a graphical user interface, the one or more labels;
where the displaying of the one or more labels includes a graphic report of at least one time period and a label that correspond to the time period.