Granted Patent B2

US 12,369,838 · App. 18/112,040 · Granted Jul 29, 2025

Inventors: Pramodsingh Hirasingh Thakur (Woodbury, MN); Rezwan Ahmed (Arden Hills, MN); Stephen B. Ruble (Lino Lakes, MN)

Assignee: Cardiac Pacemakers, Inc.

A61B5/361A61B5/1135A61B5/74G16H10/60G16H50/30

Abstract

Claims (27)

1. A medical-device system for monitoring a patient at a risk of cardiac arrhythmia, comprising:

a risk stratifier circuit configured to, in an absence of prior and present atrial arrhythmia, determine a composite risk of the patient developing a future atrial arrhythmia using a trained machine-learning model and a plurality of features of physiological information sensed via multiple sensors associated with the patient; and

an arrhythmia monitor circuit configured to adjust an arrhythmia monitoring parameter including to selectively activate or deactivate one or more of the multiple sensors to acquire or process sensor data based at least in part on the composite risk, and to detect an atrial arrhythmia event using the adjusted arrhythmia monitoring parameter.

2. The medical-device system of claim 1 , wherein the arrhythmia monitor circuit is configured to, in response to the composite risk exceeding a risk threshold, automatically adjust the arrhythmia monitoring parameter to increase a sensitivity for detecting the atrial arrhythmia event.

3. The medical-device system of claim 2 , wherein the arrhythmia monitoring parameter being adjusted to increase the sensitivity for detecting the atrial arrhythmia event includes an arrhythmia detection threshold.

4. The medical-device system of claim 2 , wherein the arrhythmia monitoring parameter being adjusted to increase the sensitivity for detecting the atrial arrhythmia event includes a model parameter of the trained machine-learning model.

5. The medical-device system of claim 1 , comprising a sensor circuit coupled to the multiple sensors and configured to collect the physiological information from the patient in accordance with a data acquisition plan,

wherein to adjust an arrhythmia monitoring parameter, the arrhythmia monitor circuit is configured to automatically adjust the data acquisition plan based at least in part on the composite risk.

6. The medical-device system of claim 5 , wherein to adjust the data acquisition plan includes to increase a data acquisition duration in response to the composite risk exceeding a risk threshold.

7. The medical-device system of claim 5 , wherein to adjust the data acquisition plan includes to increase a data acquisition frequency in response to the composite risk exceeding a risk threshold.

8. The medical-device system of claim 1 , wherein the risk stratifier circuit is configured to, in response to the composite risk exceeding a risk threshold, adaptively update a subsequent determination of the composite risk in the patient.

9. The medical-device system of claim 8 , wherein to adaptively update the subsequent determination of the composite risk includes to update the features of physiological information being applied to the trained machine-learning model.

10. The medical-device system of claim 8 , wherein to adaptively update the subsequent determination of the composite risk includes to adjust a model parameter of the trained machine-learning model.

11. The medical-device system of claim 1 , comprising a model training circuit configured to generate the trained machine-learning model using physiological information sensed from a patient population.

12. The medical-device system of claim 1 , wherein the plurality of features include one or more heart sound features, one or more thoracic impedance features, or one or more respiration features.

13. The medical-device system of claim 1 , comprising a therapy circuit configured to generate or adjust a antiarrhythmic therapy in accordance with the composite risk.

14. A method of monitoring a patient at a risk of cardiac arrhythmia, the method comprising:

receiving physiological information sensed via multiple sensors associated with the patient;

in an absence of prior and present atrial arrhythmia, determining a composite risk of the patient developing a future atrial arrhythmia using a trained machine-learning model and a plurality of features of the received physiological information;

via an arrhythmia monitor circuit, adjusting an arrhythmia monitoring parameter including selectively activating or deactivate one or more of the multiple sensors to acquire or process sensor data based at least in part on the composite risk; and

detecting an atrial arrhythmia event using the adjusted arrhythmia monitoring parameter.

15. The method of claim 14 , wherein adjusting the arrhythmia monitoring parameter includes increasing a sensitivity for detecting the atrial arrhythmia event in response to the composite risk exceeding a risk threshold.

16. The method of claim 15 , wherein the arrhythmia monitoring parameter being adjusted to increase the sensitivity for detecting the atrial arrhythmia event includes at least one of an arrhythmia detection threshold or a model parameter of the trained machine-learning model.

17. The method of claim 14 , wherein adjusting the arrhythmia monitoring parameter includes automatically adjusting a data acquisition plan for collecting the physiological information from the patient based at least in part on the composite risk.

18. The method of claim 17 , wherein adjusting the data acquisition plan includes increasing a data acquisition duration or a data acquisition frequency in response to the composite risk exceeding a risk threshold.

19. The method of claim 14 , comprising, in response to the composite risk exceeding a risk threshold, adaptively updating a subsequent determination of the composite risk in the patient, including adjusting a model parameter of the trained machine-learning model or updating the features of physiological information.

20. The method of claim 14 , comprising generating or adjusting an antiarrhythmic therapy in accordance with the composite risk.

Continuity (3)

References Cited (45)