Sleep quality estimation device, sleep quality estimation method and program for sleep quality estimation

1. A sleep quality estimation device comprising:

a respiration sensor that detects a respiratory waveform of a subject;

a processor configured to:

estimate a depth of sleep of the subject during sleep based on the detected respiratory waveform of the subject,

estimate an appearance time of δ-wave of the subject during sleep from the estimated depth of sleep of the subject, and

estimate a sleep quality of the subject based on an accumulated appearance time of δ-wave;

a database in which models associated with an appearance probability of δ-wave for the depth of sleep during sleep are recorded,

wherein the processor is configured to estimate the appearance time of δ-wave of the subject during sleep based on the depth of sleep of the subject during sleep and the models associated with the appearance probability of δ-wave for the depth of sleep during sleep recorded in the database and to estimate the sleep quality of the subject based on the appearance time of δ-wave,

wherein the models recorded in the database are associated with the appearance probability of δ-wave for a stay time when the depth of sleep of each sleep cycle from non-REM sleep to REM sleep stays with a value deeper than a predetermined value, and

the processor estimates the stay time of each sleep cycle from the depth of sleep of the subject during sleep and estimates the appearance time of δ-wave of the subject during sleep based on the stay time and the models associated with the appearance probability of δ-wave for the stay time of each sleep cycle,

wherein, in the models recorded in the database, the longer the elapsed time from the onset of sleep is, the lower the appearance probability of δ-wave for the sleep depth is set.

2. A sleep quality estimation device comprising:

a respiration sensor that detects a respiratory waveform of a subject;

a processor configured to:

estimate a depth of sleep of the subject during sleep based on the detected respiratory waveform of the subject,

acquire a time series information related to the estimated depth of sleep of the subject during sleep,

estimate an appearance time of δ-wave of the subject during sleep from the depth of sleep of the subject during sleep according to the time series information,

estimates a sleep quality of the subject based on an accumulated appearance time of δ-wave;

a database in which models associated with an appearance probability of δ-wave for the depth of sleep during sleep are recorded,

wherein the processor is configured to estimate the appearance time of δ-wave of the subject during sleep based on the depth of sleep of the subject during sleep and the models associated with the appearance probability of δ-wave for the depth of sleep during sleep recorded in the database and to estimate the sleep quality of the subject based on the appearance time of δ-wave,

wherein the models recorded in the database are associated with the appearance probability of δ-wave for a stay time when the depth of sleep of each sleep cycle from non-REM sleep to REM sleep stays with a value deeper than a predetermined value, and

the processor estimates the stay time of each sleep cycle from the depth of sleep of the subject during sleep and estimates the appearance time of δ-wave of the subject during sleep based on the stay time and the models associated with the appearance probability of δ-wave for the stay time of each sleep cycle,

wherein, in the models recorded in the database, the longer the elapsed time from the onset of sleep is, the lower the appearance probability of δ-wave for the sleep depth is set.

3. A sleep quality estimation method comprising:

detecting, using a respiration sensor, a respiratory waveform of a subject;

estimating, using a processor, a depth of sleep of the subject during sleep based on the detected respiratory waveform;

estimating, using the processor, an appearance time of δ-wave of the subject during sleep from the estimated depth of sleep of the subject;

estimating, using the processor, a sleep quality of the subject based on an accumulated appearance time of δ-wave;

providing a database in which models associated with an appearance probability of δ-wave for the depth of sleep during sleep are recorded;

wherein, in the sleep quality estimating, the appearance time of δ-wave of the subject during sleep is estimated based on the depth of sleep of the subject during sleep and the models associated with the appearance probability of δ-wave for the depth of sleep during sleep recorded in the database;

wherein the models recorded in the database are associated with the appearance probability of δ-wave for a stay time when the depth of sleep of each sleep cycle from non-REM sleep to REM sleep stays with a value deeper than a predetermined value, and

in the sleep quality estimating, the stay time of each sleep cycle is estimated from the depth of sleep of the subject during sleep and the appearance time of δ-wave of the subject during sleep is estimated based on the models associated with the appearance probability of δ-wave for the stay time of each sleep cycle,

wherein, in the models recorded in the database, the longer the elapsed time from the onset of sleep is, the lower the appearance probability of δ-wave for the sleep depth is set.

4. A sleep quality estimation method comprising:

detecting, using a respiration sensor, a respiratory waveform of a subject;

estimating, by a processor, a depth of sleep of the subject during sleep based on the detected respiratory waveform of the subject;

acquiring, by the processor, a time series information related to the estimated depth of sleep of the subject during sleep;

estimating, by the processor, the appearance time of δ-wave of the subject during sleep from the depth of sleep of the subject during sleep according to the time series information;

estimating, by the processor, a sleep quality of the subject based on an accumulated appearance time of δ-wave;

providing a database in which models associated with an appearance probability of δ-wave for the depth of sleep during sleep are recorded,

wherein, in the sleep quality estimating, the appearance time of δ-wave of the subject during sleep is estimated based on the depth of sleep of the subject during sleep and the models associated with the appearance probability of δ-wave for the depth of sleep during sleep recorded in the database and the sleep quality of the subject is estimated based on the accumulated appearance time of δ-wave,

wherein the models recorded in the database are associated with the appearance probability of δ-wave for a stay time when the depth of sleep of each sleep cycle from non-REM sleep to REM sleep stays with a value deeper than a predetermined value, and

in the sleep quality estimating, the stay time of each sleep cycle is estimated from the depth of sleep of the subject during sleep and the appearance time of δ-wave of the subject during sleep is estimated based on the models associated with the appearance probability of δ-wave for the stay time of each sleep cycle,

wherein, in the models recorded in the database, the longer the elapsed time from the onset of sleep is, the lower the appearance probability of δ-wave for the sleep depth is set.

5. A program for sleep quality estimation which causes a computer to execute:

a sleep stage information acquisition step of acquiring time series information relating to a depth of sleep of a subject during sleep; and

a sleep quality estimation step of estimating an appearance time of δ-wave of the subject during sleep from the depth of sleep of the subject during sleep according to the time series information acquired in the sleep stage information acquisition step and estimating a sleep quality of the subject based on an accumulated appearance time of δ-wave,

wherein the sleep quality estimation step causes the computer to estimate the appearance time of δ-wave of the subject during sleep based on the depth of sleep of the subject during sleep and models associated with an appearance probability of δ-wave for the depth of sleep during sleep recorded in a database and causes the computer to estimate the sleep quality of the subject based on an accumulated appearance time of δ-wave;

wherein the models recorded in the database are associated with the appearance probability of δ-wave for a stay time when the depth of sleep of each sleep cycle from non-REM sleep to REM sleep stays with a value deeper than a predetermined value, and

the sleep quality estimation step causes the computer to estimate the stay time of each sleep cycle from the depth of sleep of the subject during sleep and causes the computer to estimate the appearance time of δ-wave of the subject during sleep based on the models associated with the appearance probability of δ-wave for the stay time of each sleep cycle,

wherein, in the models recorded in the database, the longer the elapsed time from the onset of sleep is, the lower the appearance probability of δ-wave for the sleep depth is set.