System and method for cardiorespiratory sleep stage classification

Patent metadata

Field	Value
Publication number	US-2017360308-A1
Application number	US-201515536680-A
Country	US
Kind code	A1
Filing date	Dec 10, 2015
Priority date	Dec 18, 2014
Publication date	Dec 21, 2017
Grant date	—

Abstract

The present disclosure pertains to a system configured to determine one or more parameters based on cardiorespiratory information from a subject and determine sleep stage classifications based on a discriminative undirected probabilistic graphical model such as Conditional Random Fields using the determined parameters. The system is advantageous because sleep is a structured process in which parameters determined for individual epochs are not independent over time and the system determines the sleep stage classifications based on parameters determined for a current epoch, determined relationships between parameters, sleep stage classifications determined for previous epochs, and/or other information. The system does not assume that determined parameters are discriminative during an entire sleep stage, but maybe indicative of a sleep stage transition alone. In some embodiments, the system comprises one or more sensors, one or more physical computer processors, electronic storage, and a user interface.

First claim

1 . A system configured to determine sleep stage indicators in a subject for individual epochs of time during a sleep session, the system comprising: one or more sensors configured to generate output signals conveying information related to cardiorespiratory activity of the subject, the cardiorespiratory activity including cardiac activity and respiratory activity; and
one or more physical computer processors configured by computer readable instructions to:
determine one or more cardiorespiratory activity parameters of the subject based on the output signals the one or more cardiorespiratory activity parameters including cardiorespiratory coupling parameters, wherein the cardiorespiratory coupling parameters indicate a strength of a coupling between cardiac and respiratory autonomic systems of the subject, and wherein said determining the one or more cardiorespiratory activity parameters comprises (a) determining a first cardiorespiratory activity parameter and a second cardiorespiratory activity parameter for a first epoch of time, and (b) determining the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter for a second epoch of time;
determine one or more relationships between the one or more cardiorespiratory activity parameters of the subject, the one or more relationships including relationships between the cardiorespiratory coupling parameters, wherein said determining the one or more relationships comprises determining a relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter for multiple individual epochs of time; and
determine sleep stage indicators that indicate a sleep stage of the subject for the individual epochs of time based on the one or more determined relationships and the one or more determined parameters such that:
a first sleep stage indicator for the first epoch of time is determined based on the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter determined for the first epoch of time and the relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter over multiple individual epochs of time; and a second sleep stage indicator for the second epoch of time is determined based on the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter determined for the second epoch of time, the relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter over multiple individual epochs of time, and the first sleep stage indicator.

2 . The system of claim 1 , wherein the one or more physical computer processors are configured such that the sleep stage indicators are probability indicators that indicate a probability of the subject being in one or more possible sleep stages for an individual epoch of time.

3 . The system of claim 1 , wherein the one or more physical computer processors are configured determine the sleep stage indicators using conditional random fields.

4 . The system of claim 1 , wherein the one or more physical computer processors are configured such that the sleep stage indicators indicate transitions between sleep stages in the subject.

5 . (canceled)

6 . A method for determining sleep stage indicators in a subject for individual epochs of time during a sleep session with a determination system, the system comprising one or more sensors and one or more physical computer processors, the method comprising:
generating, with the one or more sensors, output signals conveying information related to cardiorespiratory activity of the subject, the cardiorespiratory activity including cardiac activity and respiratory activity;
determining, with the one or more physical computer processors, one or more cardiorespiratory activity parameters of the subject based on the output signals, the one or more cardiorespiratory activity parameters including cardiorespiratory coupling parameters, wherein the cardiorespiratory coupling parameters indicate a strength of a coupling between cardiac and respiratory autonomic systems of the subject, and wherein said determining the one or more cardiorespiratory activity parameters comprises (a) determining a first cardiorespiratory activity parameter and a second cardiorespiratory activity parameter for a first epoch of time and (b) determining the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter for a second epoch of time;

determining, with the one or more physical computer processors, one or more relationships between the one or more cardiorespiratory activity parameters of the subject, the one or more relationships including relationships between the cardiorespiratory coupling parameters wherein said determining the one or more relationships comprises determining a relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter for multiple individual epochs of time; and
determining, with the one or more physical computer processors, sleep stage indicators that indicate a sleep stage of the subject for the individual epochs of time based on the one or more determined relationships and the one or more determined parameters such that:
a first sleep stage indicator for the first epoch of time is determined based on the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter determined for the first epoch of time and the relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter over multiple individual epochs of time; and
a second sleep stage indicator for the second epoch of time is determined based on the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter determined for the second epoch of time, the relationship between the first cardiorespiratory activity parameter and the second cardiorespiratory activity parameter over multiple individual epochs of time, and the first sleep stage indicator.

7 . The method of claim 6 , wherein the sleep stage indicators are probability indicators that indicate a probability of the subject being in one or more possible sleep stages for an individual epoch of time.

8 . The method of claim 6 , wherein the sleep stage indicators are determined using conditional random fields.

9 . The method of claim 6 , wherein the sleep stage indicators indicate transitions between sleep stages in the subject.

10 . (canceled)

11 . A system configured to determine sleep stage indicators in a subject for individual epochs of time during a sleep session, the system comprising: means for generating output signals conveying information related to cardiorespiratory activity of the subject the cardiorespiratory activity including cardiac activity and respiratory activity;
means for determining one or more cardiorespiratory activity parameters of the subject based on the output signals, the one or more cardiorespiratory activity parameters including cardiorespiratory coupling parameters, wherein the cardiorespiratory coupling parameters indicate a strength of a coupling between cardiac and respiratory autonomic systems of the subject, and wherein said determining the one or more cardiorespiratory activity parameters comprises (a) determining a first cardiorespiratory activity parameter and a second cardiorespiratory activity parameter for a first epoch of time and (b) determining the first cardiorespiratory activity parameter and the second cardiores

Assignees

• Koninklijke Philips Nv

Inventors

• Fonseca Pedro Miguel
• Long Xi
• Den Teuling Nicolaas Gregorius Petrus
• Haakma Reinder
• Aarts Ronaldus Maria

Classifications

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• A61B5/4812

  Detecting sleep stages or cycles · CPC title

• A61B5/4806 Primary

  Sleep evaluation (A61B5/4821 takes precedence; devices for inducing sleep A61M21/02) · CPC title

• A61B5/024 Primary

  Measuring pulse rate or heart rate · CPC title

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title