US 12,451,245 B2
Prediction of usage or compliance
Nathan Zersee Liu, Sydney (AU); Sakeena De Souza, Sydney (AU); and Oleksandr Gromenko, Singapore (SG)
Assigned to ResMed Pty Ltd, Bella Vista (AU)
Appl. No. 17/419,283
Filed by ResMed Pty Ltd, Bella Vista (AU)
PCT Filed Dec. 23, 2019, PCT No. PCT/US2019/068375
§ 371(c)(1), (2) Date Jun. 28, 2021,
PCT Pub. No. WO2020/139845, PCT Pub. Date Jul. 2, 2020.
Claims priority of provisional application 62/786,225, filed on Dec. 28, 2018.
Prior Publication US 2022/0076822 A1, Mar. 10, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G16H 40/67 (2018.01); A61M 16/06 (2006.01); G16H 10/60 (2018.01); G16H 20/40 (2018.01); G16H 50/30 (2018.01); A61M 16/16 (2006.01)
CPC G16H 40/67 (2018.01) [A61M 16/0616 (2014.02); G16H 10/60 (2018.01); G16H 20/40 (2018.01); G16H 50/30 (2018.01); A61M 16/16 (2013.01); A61M 2202/0208 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A system for predicting compliance, the system comprising:
a device configured to output usage data, the device including a respiratory therapy device;
a memory containing machine readable medium comprising machine executable code having instructions stored thereon; and
a control system coupled to the memory and comprising one or more processors, the control system configured to execute the machine executable code to cause the control system to:
receive, from the device, a set of usage data for a session for a user;
identify a set of previously stored usage data for the user within a past time window, the past time window being split into a first duration and a second duration;
process the set of usage data and the set of previously stored usage data with an algorithm to determine a likelihood the user is to reduce usage of the device within a future time window, the algorithm including a regression model with respective coefficients scaling respective inputs to the regression model such that the regression model combines the scaled respective inputs to determine the likelihood the user is to reduce usage, the regression model having a plurality of features as inputs, the features including an average hours of usage of the device during the first duration and an average hours of usage of the device during the second duration;
determine the user is likely to reduce usage based on if the determined likelihood being above a predetermined threshold; and
adjust a setting on the device for lowering the likelihood, the setting being determined from an identified issue associated with the user and the device, wherein adjusting the setting includes (i) reducing an initial pressure on a ramp feature of the respiratory therapy device, and/or (ii) increasing a humidification level associated with the respiratory therapy device.