Systems and methods for using breath events in sleep staging

A method comprising:

• generating, with one or more sensors, output signals conveying information related to one or more breathing parameters of subject;
• determining, with one or more physical computer processors, based on the output signals, one or more breathing features of individual breaths of the subject;
• determining, with one or more physical computer processors, at least a first meta-feature from the one or more breathing features over a number of first time windows each having a first size and a second meta-feature from the one or more breathing features over a number of second time windows each having a second size different than the first size, wherein the first meta-feature is a first statistic that describes a first shape of a first distribution of the one or more breathing features over the number of first time windows and is not just a variance or standard deviation of the one or more breathing features over time, and wherein the second meta-feature is a second statistic obtained from the one or more breathing features that describes a shape of a distribution of the one or more breathing features over the number of second time windows and is not just the variance or standard deviation of the one or more breathing features over time;
• detecting, with one or more physical computer processors, presence of one or more breathing events based on the output signals and indicating a category of a type of each of the one or more breathing events;
• and
• determining, with one or more physical computer processors, sleep states of the subject with a sleep stage classifier model based upon the first meta-feature, the second meta-feature, and the category of each of the one or more breathing events, wherein the sleep stage classifier model is based on a history of prior sleep states and is not a Markov classification system.

A sleep staging system comprising:

• means for sensing, configured for generating output signals conveying information related to one or more breathing parameters of subject;
• means for determining, configured for determining, based on the output signals, one or more breathing features of individual breaths of the subject;
• means for determining, configured for determining at least a first meta-feature from the one or more breathing features over a number of first time windows each having a first size and a second meta-feature from the one or more breathing features over a number of second time windows each having a second size different than the first size, wherein the first meta-feature is a first statistic that describes a first shape of a first distribution of the one or more breathing features over the number of first time windows and is not just a variance or standard deviation of the one or more breathing features over time, and wherein the second meta-feature is a second statistic obtained from the one or more breathing features that describes a shape of a distribution of the one or more breathing features over the number of second time windows and is not just the variance or standard deviation of the one or more breathing features over time;
• means for detecting, configured for detecting presence of one or more breathing events based on the output signals and indicating a category of a type of each of the one or more breathing events;
• and
• means for determining, configured for determining sleep states of the subject with a sleep stage classifier model based upon the first meta-feature, the second meta-feature, and the category of each of the one or more breathing events, wherein the sleep stage classifier model is based on a history of prior sleep states and is not a Markov classification system.