Heart failure event detection using multi-level categorical fusion

• US 10,182,768 B2
• Filed: 04/18/2017
• Issued: 01/22/2019
• Est. Priority Date: 03/07/2014
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Zhang, Yi, Thakur, Pramodsingh Hirasingh, An, Qi, Thompson, Julie A., Wariar, Ramesh, Sweeney, Robert J.
• Primary Examiner(s): Flory, Christopher A
• Application Number:
  US15/490,226
• Publication Number:
  US 20170215809A1
• Time in Patent Office: 644 Days
• Field of Search: 600301, 600481, 600483, 600508, 600509, 600529
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/4842 Monitoring progression or stage of a disease
• CPC Class Codes: A61B 5/686 Permanently implanted devices, e.g. pacemakers, other stimulators, biochips A61B5/6861 takes precedence
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 7/00 Instruments for auscultation
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Systems and methods for detecting a present, or predicting a future, target physiologic event such as worsening heart failure (HF) are described. A system can comprise a patient information receiver circuit, at least two categorical risk analyzer circuits, and a categorical fusion circuit. The patient information receiver circuit receives physiologic signals and generates signal trends. The categorical risk analyzer circuit receives a category-specific input selected from the signal trends according to an associative physiologic condition indicative or correlative of the target event. The categorical risk analyzer circuit produces a signal trend metric indicating relative change in signal strength over time. The categorical risk analyzer circuit calculates a categorical risk index that indicates likelihood of the patient developing or presenting the associative physiologic condition. The categorical fusion circuit uses the categorical risk indices to generate a composite risk indicator indicating likelihood of the patient developing the target physiologic event.

Claims

1. A system for detecting a target physiological event, comprising:

a receiver circuit configured to receive patient physiological data;

at least two categorical risk analyzer circuits, each including;

a category-specific input selection circuit configured to select a portion of the received physiological data according to an associative physiological condition, wherein the associative physiological condition is a different type of condition than the target physiological event; and

a categorical risk calculator circuit configured to calculate a categorical risk index using the selected portion of the received physiological data, the categorical risk index indicating a patient risk of having the associative physiological condition; and

a detector circuit, coupled to the at least two categorical risk analyzer circuits, configured to generate a composite risk indicator using the categorical risk indices respectively provided by the at least two categorical risk analyzer circuits, and to detect the target physiological event using the composite risk indicator.

2. The system of claim 1, wherein the categorical risk calculator circuit is configured to calculate the categorical risk index using a trend of a signal feature determined from the selected portion of the received physiological data.

3. The system of claim 1, wherein the categorical risk calculator circuit is configured to calculate the categorical risk index using a signal change of the selected portion of the received physiological data over time.

4. The system of claim 1, wherein the categorical risk calculator circuit is configured to calculate the categorical risk index using a signal strength of the selected portion of the received physiological data.

5. The system of claim 1, wherein the categorical risk calculator circuit is configured to calculate the categorical risk index using a signal variability of the selected portion of the received physiological data.

6. The system of claim 1, wherein the target physiological event includes a worsening heart failure (WHF) event, and the associative physiological conditions include comorbidity conditions, and wherein:

the categorical risk calculator circuits of at least two categorical risk analyzer circuits are configured to calculate respective categorical risk indices corresponding to the comorbidity conditions; and

the detector circuit is configured to generate a composite risk indicator indicating an elevated risk of WHF using the categorical risk indices corresponding to the comorbidity conditions, and to detect the WHF using the composite risk indicator.

7. The system of claim 1, wherein the associative physiological condition includes a symptomatic cardiac event or a symptomatic pulmonary event.

8. The system of claim 1, wherein the associative physiological condition includes a specified etiologic factor of the target physiological event.

9. The system of claim 1, further comprising a therapy circuit for delivering a therapy based on the composite risk indicator.

10. A non-transitory machine-readable storage medium, comprising a plurality of instructions that, responsive to being executed with processor circuitry of a computing device, cause the computing device to:

receive patient physiological data obtained;

select a first portion of the received physiological data according to a first associative physiological condition, and select a second portion of the received physiological data according to a second associative physiological condition, the first and second associative physiological conditions each being of a different type of condition than a target physiological event;

calculate a first categorical risk index using the selected first portion of the received physiological data, and calculating a second categorical risk index using the selected second portion of the received physiological data, wherein the first categorical risk index indicates a patient risk of having the first associative physiological condition; and

the second categorical risk index indicates a patient risk of having the second associative physiological condition;

generate a composite risk indicator using the first and second categorical risk indices; and

detect the target physiological event using composite risk indicator.

11. The non-transitory machine-readable storage medium of claim 10, wherein the instructions to calculate the first and second categorical risk indices include instructions that cause the computing device to:

calculate the first categorical risk index using a first signal feature trend generated from the selected first portion of the received physiological data; and

calculate the second categorical risk index using a second signal feature trend generated from the selected second portion of the received physiological data.

12. The non-transitory machine-readable storage medium of claim 10, wherein the instructions to select the first and second portions of the received physiological data include instructions that cause the computing device to adaptively update the selection of at least one of the first or second data portion in response to one or more of a change in patient context;

a change in environmental context, or an update of domain knowledge.

13. The non-transitory machine-readable storage medium of claim 10, wherein the instructions to calculate the first and second categorical risk indices include instructions that cause the computing device to determine the first and second categorical risk indices as categorical decisions or numerical risk quantities.

14. The non-transitory machine-readable storage medium of claim 10, wherein the instructions to generate a composite risk indicator include instructions that cause the computing device to determine a linear or a nonlinear combination of the first and second categorical risk indices.

15. A method for detecting a target physiological event via a medical system, the method comprising:

receiving patient physiological data via a receiver circuit;

selecting;

via an input selection circuit;

a first portion of the received physiological data according to a first associative physiological condition, and a second portion of the received physiological data according to a second associative physiological condition, the first and second associative physiological conditions each being of a different type of condition than a target physiological event;

calculating, via a risk analyzer circuit, a first categorical risk index using the selected first portion of the received physiological data, and a second categorical risk index using the selected second portion of the received physiological data, wherein the first categorical risk index indicates a patient risk of having the first associative physiological condition, and the second categorical risk index indicates a patient risk of having the second associative physiological condition;

generating, via a detector circuit, a composite risk indicator using the first and second categorical risk indices; and

detecting, via the detector circuit, the target physiological event using composite risk indicator.

16. The method of claim 15, further comprising adaptively updating the selection of at least one of the first or second data portion in response to one or more of a change in patient context, a change in environmental context, or an update of domain knowledge.

17. The method of claim 15, wherein e first and second associative physiological conditions each including one or more of:

a patient symptom;

an etiologic factor of the target physiological event;

a patient-specific risk stratifier;

a patient comorbidity condition;

or;

a patient vital sign.

18. The method of claim 15, further comprising receiving at least one of patient demographic information or patient therapy information;

wherein calculating the first or second categorical risk index includes calculating the respective risk index using the at least one of the patient demographic information or the patient therapy information.

19. The method of claim 15, wherein the target physiological event includes a worsening heart failure (WHF) event, and the first and second associative physiological conditions include respectively first and second comorbidity conditions, and wherein:

calculating the first and second categorical risk indices includes the first categorical risk index corresponding to the first comorbidity condition, and the second categorical risk index corresponding to the second comorbidity condition; and

generating the composite risk indicator includes generating an indicator of an elevated risk of WHF using the first and second categorical risk indices.

20. The method of claim 19, wherein generating the composite risk indicator includes determining a linear or a nonlinear combination of the first and second categorical risk indices.