Sleep estimation device, sleep estimation system, wearable instrument, and sleep estimation method

1 . A sleep estimation device comprising:

a processor that executes the following functions:

acquiring blood flow data indicating a blood flow of a subject;

generating a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data; and

determining a stage of sleep of the subject based on the frequency spectrum, wherein

when the stage of sleep in non-REM sleep is classified into stages 1 to 3 comprising a stage 1, a stage 2 and a stage 3, in an order from a lightest stage of sleep,

the processor determines that the stage of sleep of the subject has transitioned from stage 1 to stage 2 or stage 3 in a case where a frequency band of 0.2 to 0.3 Hz is divided into a first range and a second range and the generated frequency spectrum has a first intensity that is a maximum intensity in the first range and a second intensity that is a maximum intensity in the second range and the first intensity is larger than the second intensity.

2 . The sleep estimation device according to claim 1 , wherein the processor is further configured to execute the function:

performing first notification processing subsequent to determining that the stage of sleep of the subject has transitioned from the stage 1 to the stage 2 or the stage 3, or subsequent to determining that the stage of sleep of the subject has transitioned from the stage 1 to the stage 2 or the stage 3 and a predetermined time has elapsed.

3 . The sleep estimation device according to claim 1 , wherein the processor is further configured to execute the functions:

generating, as the frequency spectrum, intensity change data indicating a temporal change in intensity in each frequency band within a predetermined time by performing time-frequency analysis processing as the frequency analysis processing, and

adding the intensity change data to a learned model, the learned model learned using teacher data in which a correct answer label is associated with the intensity change data known to correspond to the stage 2 or the stage 3, thereby determining that the stage of sleep of the subject has transitioned from the stage 1 to the stage 2 or the stage 3.

4 . The sleep estimation device according to claim 1 , wherein the processor is further configured to execute the functions:

acquiring acceleration data indicating an acceleration caused by a movement of the subject; and

determining whether the subject is stationary based on the acceleration data, wherein

the frequency analysis processing is performed when the subject is stationary.

5 . A sleep estimation device comprising:

a blood flow meter comprising a light emitting unit configured to irradiate a blood vessel of the subject with light, and a light receiving unit configured to receive scattered light generated by the irradiation;

a processor that executes the following functions:

acquiring blood flow data from the blood flow meter, the blood flow data indicating a blood flow of a subject based on the received scattered light;

generating a frequency spectrum indicating a result of time-frequency analysis processing on the blood flow data by performing wavelet transform processing or short-time Fourier transform processing, in which an intensity in a predetermined frequency band is relatively emphasized compared with other frequency bands, on the blood flow data; and

determining a stage of sleep of the subject based on the frequency spectrum, wherein

when the stage of sleep in non-REM sleep is classified into stages 1 to 3 comprising a stage 1, a stage 2 and a stage 3, in an order from a lightest stage of sleep,

the processor determines that the stage of sleep of the subject has transitioned from stage 1 to stage 2 or stage 3 in a case where a frequency band of 0.2 to 0.3 Hz is divided into a first range and a second range and the generated frequency spectrum has a first intensity that is a maximum intensity in the first range and a second intensity that is a maximum intensity in the second range and the first intensity is larger than the second intensity.

6 . The sleep estimation device according to claim 5 , wherein the processor is further configured to execute the functions:

generating, as the frequency spectrum, intensity change data indicating a temporal change in intensity in each frequency band within a predetermined time, and

adding the intensity change data to a learned model, the learned model learned using teacher data in which a correct answer label is associated with the intensity change data known to correspond to the stage 2 or the stage 3, thereby determining that the stage of sleep of the subject has transitioned from the stage 1 to the stage 2 or the stage 3.

7 . A sleep estimation system comprising:

a portable terminal including the sleep estimation device according to claim 1 ; and

a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating a blood vessel of the subject with light.

8 . A wearable instrument comprising:

the sleep estimation device according to claim 1 ; and

a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating a blood vessel of the subject with light.

9 . A sleep estimation method comprising:

detecting blood flow data by irradiating a blood vessel of a subject with light and receiving scattered light generated by the irradiation, wherein the blood flow data indicates a blood flow of the subject;

generating a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data;

determining a stage of sleep of the subject based on the frequency spectrum, wherein when the stage of sleep in non-REM sleep is classified into stages 1 to 3 comprising a stage 1, a stage 2 and a stage 3, in an order from a lightest stage of sleep;

dividing a frequency band of 0.2 to 0.3 Hz into a first range and a second range;

determining a first intensity that is a maximum intensity in the first range and a second intensity that is a maximum intensity in the second range; and

determining that the stage of sleep of the subject has transitioned from stage 1 to stage 2 or stage 3 based on the first intensity being larger than the second intensity.

10 . A sleep estimation method comprising:

acquiring blood flow data indicating a blood flow of a subject;

generating a frequency spectrum indicating a result of time-frequency analysis processing on the blood flow data by performing wavelet transform processing or short-time Fourier transform processing, in which an intensity in a predetermined frequency band is relatively emphasized compared with other frequency bands, on the blood flow data;

determining a stage of sleep of the subject based on the frequency spectrum, wherein when the stage of sleep in non-REM sleep is classified into stages 1 to 3 comprising a stage 1, a stage 2 and a stage 3, in an order from a lightest stage of sleep;

dividing a frequency band of 0.2 to 0.3 Hz into a first range and a second range;

determining a first intensity that is a maximum intensity in the first range and a second intensity that is a maximum intensity in the second range; and

determining that the stage of sleep of the subject has transitioned from stage 1 to stage 2 or stage 3 based on the first intensity being larger than the second intensity.