Respiration monitoring optimization method, computer device, and storage medium

Abstract

Disclosed are a respiration monitoring optimization method, a computer device, and a storage medium. The method comprises: acquiring first indicator values respectively corresponding to at least two physiological indicators, wherein the first indicator values are estimated on the basis of first PPG signals sampled by a PPG sensor under the control of a sampling controller; predicting and obtaining, on the basis of the first indicator values, predicted values of the physiological indicators in a next respiratory cycle; determining, on the basis of the first indicator values and the predicted values, target sampling parameters for the next respiratory cycle required when the estimation accuracy is the highest; and adjusting, on the basis of the target sampling parameters, sampling parameters of the sampling controller, and during sampling in the next respiratory cycle, obtaining, on the basis of the adjusted sampling parameters, third indicator values of the physiological indicators, and updating, on the basis of the third indicator values, prediction function parameters. The prediction function parameters comprise calculation parameters corresponding to blood oxygen, heart rate, and respiration.

Description

Respiration monitoring optimization method, computer device and storage medium

[0001] Related applications

[0002] The present application claims priority to Chinese Patent Application No. 202411283126.4, filed on September 13, 2024, entitled "Respiration Monitoring Precision Optimization Method, Device, Computer Device and Storage Medium", Chinese Patent Application No. 202411282931.5, filed on September 13, 2024, entitled "Respiration Monitoring Power Consumption Optimization Method, Device, Computer Device and Storage Medium", Chinese Patent Application No. 202411283666.2, filed on September 13, 2024, entitled "Respiration Monitoring Joint Optimization Method, Computer Device and Readable Storage Medium", all of which are incorporated herein by reference in their entirety. TECHNICAL FIELD

[0003] The present application relates to the field of physiological index monitoring, and in particular to a respiration monitoring optimization method, a computer device and a storage medium. BACKGROUND

[0004] With the development of physiological index monitoring technology, sampling parameter optimization technology has emerged, which mainly optimizes sampling parameters through physiological index values.

[0005] In related technologies, after estimating the physiological index values from the signals collected by the sensor, the sampling parameters are optimized according to the characteristics of the index values.

[0006] However, only single-dimensional physiological index (such as blood oxygen) precision optimization can be performed at present, and the optimization effect is poor, resulting in low precision of the estimated index values after subsequent sampling. SUMMARY

[0007] According to various embodiments of the present application, a respiration monitoring optimization method, a computer device and a storage medium are provided.

[0008] In a first aspect, the present application provides a respiration monitoring precision optimization method, comprising:

[0009] Obtaining first index values corresponding to at least two physiological indexes respectively; each of the first index values is estimated based on a first PPG signal obtained by sampling a PPG sensor controlled by a sampling controller;

[0010] Predicting a predicted value of each of the physiological indexes in a next breathing cycle based on each of the first index values;

[0011] determine a target sampling parameter of a next breathing cycle at a highest estimation accuracy based on the first index values and the prediction values; the estimation accuracy is an accuracy of estimating second index values of the next breathing cycle based on a second PPG signal corresponding to the next breathing cycle;

[0012] adjust the sampling parameter of the sampling controller based on the target sampling parameter, and obtain third index values of the physiological indexes based on the adjusted sampling parameter when sampling in the next breathing cycle, and update the prediction function parameters based on the third index values; the prediction function parameters include calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0013] In one of the embodiments, the first index values include a first heart rate value, a first blood oxygen content and a first respiration characteristic value, the first respiration characteristic value includes a first time interval of the respiration movement in a relatively static state, and the target sampling parameter further includes a plurality of sampling time points; the determination of the target sampling parameter of the next breathing cycle at the highest estimation accuracy based on the first index values and the prediction values includes:

[0014] determining the sampling time points based on an end time point of the first time interval, a sampling value at the end time point and the first blood oxygen content, with an error estimation lower bound as an optimization target.

[0015] determining a sampling frequency of the next breathing cycle at the highest estimation accuracy and an emission intensity of an emission unit of the PPG sensor corresponding to each of the sampling time points based on the first index values and the prediction values.

[0016] In one of the embodiments, the prediction values at least include a second time interval of the respiration movement in a relatively static state in the next breathing cycle, and the sampling time points at least include a first time point corresponding to a maximum signal intensity of the next breathing cycle and at least one second time point within the second time interval.

[0017] In one of the embodiments, the first index values include a first heart rate value, a first blood oxygen content and a first respiration characteristic value, the first respiration characteristic value includes a first respiration signal characteristic value and a first time interval of the respiration movement in a relatively static state, and the prediction values of the physiological indexes in the next breathing cycle based on the first index values include:

[0018] predicting a second respiration signal characteristic value of the next breathing cycle based on the first heart rate value, the first blood oxygen content and the first respiration signal characteristic value;

[0019] predicting a second blood oxygen content of the next breathing cycle based on the first blood oxygen content and the second respiration signal characteristic value;

[0020] predict a second time interval of the respiratory motion in a relatively static state of a next respiratory cycle based on the first time interval and the first heart rate value.

[0021] In one of the embodiments, the method further comprises:

[0022] obtaining a second PPG signal sampled by the PPG sensor in the next respiratory cycle, and the second time interval of the respiratory motion in the relatively static state predicted;

[0023] determining a third time interval of the next respiratory cycle other than the second time interval;

[0024] filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators;

[0025] estimating second indicator values of the physiological indicators based on the filtered second PPG signal.

[0026] In one of the embodiments, the filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators comprises:

[0027] if there are two target physiological indicators corresponding to the second time interval in the at least two physiological indicators, determining a correlation between each sampling data corresponding to the second time interval in the second PPG signal and each of the target physiological indicators respectively;

[0028] filtering each sampling data based on the prior parameter corresponding to the target physiological indicator with the highest correlation.

[0029] In one of the embodiments, the second indicator values comprise a second heart rate value, a second blood oxygen content, and a second respiratory characteristic value, and the estimating the second indicator values of the physiological indicators based on the filtered second PPG signal comprises:

[0030] extracting a respiratory characteristic PPG signal corresponding to respiration, a blood oxygen characteristic PPG signal corresponding to blood oxygen, and a heart rate characteristic PPG signal corresponding to heart rate from the filtered second PPG signal respectively;

[0031] determining the second respiratory characteristic value based on the respiratory characteristic PPG signal; the second respiratory characteristic value comprises a second respiratory signal characteristic value and a fourth time interval of the respiratory motion in a relatively static state;

[0032] determining the second blood oxygen content based on the fourth time interval and the blood oxygen characteristic PPG signal;

[0033] determine the second heart rate value based on the fourth time interval and the heart rate feature PPG signal.

[0034] In a second aspect, the present application provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0035] obtain first index values corresponding to at least two physiological indexes respectively; each first index value is estimated based on a first PPG signal obtained by sampling a PPG sensor under the control of a sampling controller;

[0036] predict a predicted value of each physiological index in a next breathing cycle based on each first index value;

[0037] determine a target sampling parameter of the next breathing cycle required when the estimation accuracy is the highest based on each first index value and each predicted value; the estimation accuracy is the accuracy when each second index value of the next breathing cycle is estimated based on a second PPG signal corresponding to the next breathing cycle;

[0038] adjust the sampling parameter of the sampling controller based on the target sampling parameter, and obtain third index values of each physiological index based on the adjusted sampling parameter when sampling in the next breathing cycle, and update a prediction function parameter based on the third index values; the prediction function parameter comprises calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0039] In a third aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0040] obtain first index values corresponding to at least two physiological indexes respectively; each first index value is estimated based on a first PPG signal obtained by sampling a PPG sensor under the control of a sampling controller;

[0041] predict a predicted value of each physiological index in a next breathing cycle based on each first index value;

[0042] determine a target sampling parameter of the next breathing cycle required when the estimation accuracy is the highest based on each first index value and each predicted value; the estimation accuracy is the accuracy when each second index value of the next breathing cycle is estimated based on a second PPG signal corresponding to the next breathing cycle;

[0043] adjust the sampling parameter of the sampling controller based on the target sampling parameter, and obtain third index values of the physiological indexes based on the adjusted sampling parameter when sampling in the next breathing cycle, and update the prediction function parameters based on the third index values; the prediction function parameters include calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0044] In a fourth aspect, the present application provides a method for optimizing the power consumption of respiratory monitoring, comprising:

[0045] obtaining first index values corresponding to a current breathing cycle and second index values corresponding to a previous breathing cycle of at least two physiological indexes, and a power consumption characteristic coefficient of a light source of a PPG sensor; each of the first index values is estimated based on a first PPG signal obtained by controlling the PPG sensor to sample; the light source includes a red light source and an infrared light source;

[0046] predicting the predicted values of the physiological indexes in the next breathing cycle based on each of the first index values;

[0047] determining the target sampling parameter of the PPG sensor in the next breathing cycle corresponding to the lowest power consumption based on the power consumption characteristic coefficient, the second index values and the predicted values;

[0048] adjusting the sampling parameter of the sampling controller based on the target sampling parameter, and obtaining third index values of the physiological indexes based on the adjusted sampling parameter when sampling in the next breathing cycle, and updating the prediction function parameters based on the third index values; the prediction function parameters include calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0049] In one of the embodiments, each of the first index values includes a first heart rate value, a first blood oxygen content and a first respiration characteristic value, the first respiration characteristic value includes a first respiration signal characteristic value and a first time interval of respiration movement in relative static state, and the predicting the predicted values of the physiological indexes in the next breathing cycle based on each of the first index values includes:

[0050] predicting a third respiration signal characteristic value in the next breathing cycle based on the first heart rate value, the first blood oxygen content and the first respiration signal characteristic value;

[0051] predicting a third blood oxygen content in the next breathing cycle based on the first blood oxygen content and the third respiration signal characteristic value;

[0052] predicting a third time interval of respiration movement in relative static state in the next breathing cycle based on the first time interval and the first heart rate value.

[0053] In one of the embodiments, the third time interval of the respiration movement in the next breathing cycle in the relatively static state is predicted based on the first time interval and the first heart rate value, and the third time interval comprises:

[0054] The first gain of the first start time and the third gain of the first end time of the first time interval are determined based on the first heart rate value respectively;

[0055] The third start time of the third time interval of the respiration movement in the next breathing cycle in the relatively static state is determined based on the first start time and the first gain;

[0056] The third end time of the third time interval of the respiration movement in the next breathing cycle in the relatively static state is determined based on the first end time and the third gain.

[0057] In one of the embodiments, the first index value of each of the at least two physiological indexes corresponding to the current breathing cycle comprises:

[0058] The first PPG signal sampled by the PPG sensor in the current breathing cycle and the fourth time interval of the respiration movement in the current breathing cycle in the relatively static state predicted in the previous breathing cycle are obtained;

[0059] The fifth time interval of the current breathing cycle except the fourth time interval is determined;

[0060] The first PPG signal is filtered in the fourth time interval and the fifth time interval based on the prior parameters corresponding to each of the physiological indexes respectively;

[0061] The first index value corresponding to each of the physiological indexes is estimated based on the filtered first PPG signal.

[0062] In one of the embodiments, the at least two physiological indexes comprise respiration, blood oxygen and heart rate, and each of the first index values comprises a first respiration feature value, a first blood oxygen content and a first heart rate value, and the estimation of the first index value corresponding to each of the physiological indexes based on the filtered first PPG signal comprises:

[0063] The respiration feature PPG signal corresponding to the respiration, the blood oxygen feature PPG signal corresponding to the blood oxygen and the heart rate feature PPG signal corresponding to the heart rate are extracted from the filtered first PPG signal respectively;

[0064] The first respiration feature value is determined based on the respiration feature PPG signal, and the first respiration feature value comprises a first respiration signal feature value and a first time interval of the respiration movement in the relatively static state;

[0065] determine the first blood oxygen content based on the first time interval and the blood oxygen characteristic PPG signal;

[0066] determine the first heart rate value based on the first time interval and the heart rate characteristic PPG signal.

[0067] In one of the embodiments, the determining the first blood oxygen content based on the first time interval and the blood oxygen characteristic PPG signal comprises:

[0068] determining a signal characteristic corresponding to the first time interval in the blood oxygen characteristic PPG signal;

[0069] determining the first blood oxygen content based on the signal characteristic.

[0070] In one of the embodiments, the determining the target sampling parameter of the next breathing cycle corresponding to the lowest power consumption of the PPG sensor based on the power consumption characteristic coefficient, the second index value and the predicted value further comprises:

[0071] updating the prior parameter corresponding to each of the physiological indexes based on the target sampling parameter and the predicted value.

[0072] In the fifth aspect, the present application further provides a computer device comprising a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0073] obtaining first index values corresponding to a current breathing cycle and second index values corresponding to a previous breathing cycle of at least two physiological indexes respectively, and a power consumption characteristic coefficient of a light source of a PPG sensor, each of the first index values being estimated based on a first PPG signal obtained by sampling control of the PPG sensor, and the light source comprising a red light source and an infrared light source;

[0074] predicting a predicted value of each of the physiological indexes in a next breathing cycle based on each of the first index values;

[0075] determining a target sampling parameter of the next breathing cycle corresponding to the lowest power consumption of the PPG sensor based on the power consumption characteristic coefficient, the second index value and the predicted value;

[0076] adjusting the sampling parameter of the sampling controller based on the target sampling parameter, obtaining third index values of each of the physiological indexes based on the adjusted sampling parameter when sampling in the next breathing cycle, and updating a prediction function parameter based on the third index values; the prediction function parameter comprises calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0077] In a sixth aspect, the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0078] obtaining first index values of at least two physiological indexes respectively corresponding to a current breathing cycle and second index values of the at least two physiological indexes respectively corresponding to a previous breathing cycle, and a power consumption characteristic coefficient of a light source of a PPG sensor; each of the first index values is estimated based on a first PPG signal obtained by sampling the PPG sensor under control of a sampling controller; the light source includes a red light source and an infrared light source;

[0079] predicting a predicted value of each of the physiological indexes in a next breathing cycle based on each of the first index values;

[0080] determining a target sampling parameter of the PPG sensor in the next breathing cycle corresponding to a lowest power consumption based on the power consumption characteristic coefficient, the second index values, and the predicted value;

[0081] adjusting the sampling parameter of the sampling controller based on the target sampling parameter, obtaining third index values of the physiological indexes based on the adjusted sampling parameter when sampling in the next breathing cycle, and updating a prediction function parameter based on the third index values; the prediction function parameter includes a calculation parameter corresponding to blood oxygen, heart rate, and respiration.

[0082] In a seventh aspect, the present application provides a respiratory monitoring joint optimization method, comprising:

[0083] obtaining first index values of at least two physiological indexes respectively corresponding to a current breathing cycle; each of the first index values is estimated based on a first PPG signal obtained by sampling a PPG sensor under control of a sampling controller;

[0084] predicting a predicted value of each of the physiological indexes in a next breathing cycle based on each of the first index values;

[0085] determining a target sampling parameter corresponding to a minimum value obtained under joint power consumption calculation and error boundary calculation based on each of the predicted values, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient, and a current characteristic coefficient of a light source of the PPG sensor; the light source includes a red light source and an infrared light source; the target sampling parameter is used for the sampling controller to control the PPG sensor to sample in the next breathing cycle;

[0086] adjusting the sampling parameter of the sampling controller based on the target sampling parameter.

[0087] In one of the embodiments, the determining the target sampling parameter corresponding to the minimum value obtained in the joint calculation of the power consumption and the error bound calculation comprises:

[0088] performing power consumption calculation based on the prediction values, and performing error bound calculation based on the prediction values;

[0089] performing joint calculation by taking the preset power consumption optimization coefficient and the current characteristic coefficient of the light source of the PPG sensor as first coefficients in the power consumption calculation, and taking the preset accuracy optimization coefficient as a second coefficient in the error bound calculation;

[0090] determining the target sampling parameter corresponding to the minimum value obtained in the joint calculation.

[0091] In one of the embodiments, the first index values each comprise a first heart rate value, a first blood oxygen content, and a first respiration characteristic value, the first respiration characteristic value comprises a first time interval of respiration movement in relative static state, the target sampling parameter comprises a plurality of sampling time points, a plurality of sampling frequencies, and a plurality of emission intensities, and the determining the target sampling parameter corresponding to the minimum value obtained in the joint calculation comprises:

[0092] determining the sampling time points based on an end time point of the first time interval, a sampling value at the end time point, and the first blood oxygen content, with the lower bound of error estimation as an optimization target;

[0093] determining the sampling frequencies corresponding to the sampling time points and the emission intensities of the emission unit of the PPG sensor.

[0094] In one of the embodiments, the prediction values each comprise a second time interval of respiration movement in a next respiration cycle, and the sampling time points each comprise a first time point corresponding to a maximum signal intensity in the next respiration cycle and at least one second time point in the second time interval.

[0095] In one of the embodiments, the first index values each comprise a first heart rate value, a first blood oxygen content, and a first respiration characteristic value, the first respiration characteristic value comprises a first respiration signal characteristic value and a first time interval of respiration movement in relative static state, and the prediction values of the physiological indexes in a next respiration cycle obtained based on the first index values each comprise:

[0096] a second respiration signal characteristic value in the next respiration cycle obtained based on the first heart rate value, the first blood oxygen content, and the first respiration signal characteristic value;

[0097] predict a second blood oxygen content of a next breathing cycle based on the first blood oxygen content and the second respiration signal feature value;

[0098] predict a second time interval of the respiration movement in relative static state of the next breathing cycle based on the first time interval and the first heart rate value.

[0099] In one of the embodiments, the method further comprises:

[0100] obtaining a second PPG signal sampled by the PPG sensor in the next breathing cycle and the predicted second time interval of the respiration movement in relative static state;

[0101] determining a third time interval of the next breathing cycle except the second time interval;

[0102] filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators;

[0103] estimating second indicator values of the physiological indicators based on the filtered second PPG signal.

[0104] In one of the embodiments, the filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators comprises:

[0105] if there are two target physiological indicators corresponding to the second time interval in the at least two physiological indicators, determining a correlation degree between each sampling data corresponding to the second time interval in the second PPG signal and each target physiological indicator respectively;

[0106] filtering each sampling data based on the prior parameter corresponding to the target physiological indicator with the highest correlation degree.

[0107] In one of the embodiments, the second indicator values comprise a second heart rate value, a second blood oxygen content and a second respiration feature value, and the estimating the second indicator values of the physiological indicators based on the filtered second PPG signal comprises:

[0108] extracting a respiration feature PPG signal corresponding to respiration, a blood oxygen feature PPG signal corresponding to blood oxygen and a heart rate feature PPG signal corresponding to heart rate from the filtered second PPG signal respectively;

[0109] determining the second respiration feature value based on the respiration feature PPG signal; the second respiration feature value comprises a second respiration signal feature value and a fourth time interval of the respiration movement in relative static state;

[0110] determining the second blood oxygen content based on the fourth time interval and the blood oxygen characteristic PPG signal;

[0111] determining the second heart rate value based on the fourth time interval and the heart rate characteristic PPG signal.

[0112] In one of the embodiments, the sampling parameter adjustment of the sampling controller based on the target sampling parameter comprises:

[0113] In the next breath cycle sampling, third index values of the physiological indexes are obtained based on the adjusted sampling parameters;

[0114] updating the prediction function parameters based on the third index values; the prediction function parameters comprise calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0115] In the eighth aspect, the present application further provides a computer device comprising a memory and a processor, the memory stores a computer program, and the processor realizes the following steps when executing the computer program:

[0116] obtaining first index values of at least two physiological indexes corresponding to a current breath cycle respectively; each of the first index values is estimated based on a first PPG signal obtained by sampling of a PPG sensor controlled by a sampling controller;

[0117] predicting a prediction value of each of the physiological indexes in a next breath cycle based on each of the first index values;

[0118] determining a target sampling parameter corresponding to a minimum value obtained under joint power consumption calculation and error bound calculation based on each of the prediction values, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient and a current characteristic coefficient of a light source of the PPG sensor; the light source comprises a red light source and an infrared light source; the target sampling parameter is used for controlling the PPG sensor to sample in the next breath cycle by the sampling controller;

[0119] adjusting the sampling parameter of the sampling controller based on the target sampling parameter.

[0120] In the ninth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to realize the following steps:

[0121] obtaining first index values of at least two physiological indexes corresponding to a current breath cycle respectively; each of the first index values is estimated based on a first PPG signal obtained by sampling of a PPG sensor controlled by a sampling controller;

[0122] predicting a prediction value of each of the physiological indexes in a next breath cycle based on each of the first index values;

[0123] determine a target sampling parameter corresponding to a minimum value obtained in a case of joint power consumption calculation and error bound calculation based on the prediction value, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient, and a current characteristic coefficient of a light source of the PPG sensor; the light source includes a red light source and an infrared light source; and the target sampling parameter is used for the sampling controller to control the PPG sensor to sample in a next breathing cycle.

[0124] adjust the sampling parameter of the sampling controller based on the target sampling parameter.

[0125] The details of one or more embodiments of the present application are presented in the following drawings and description to make other features, objects and advantages of the present application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0126] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor based on the disclosed drawings.

[0127] FIG. 1 is a flowchart of a breathing monitoring accuracy optimization method in an embodiment.

[0128] FIG. 2 is a relationship diagram of breathing movement and PPG reflection signal intensity change over time in an embodiment.

[0129] FIG. 3 is a distribution diagram of predicted sampling time points in an embodiment.

[0130] FIG. 4 is a breathing movement waveform diagram in an embodiment.

[0131] FIG. 5 is a waveform diagram of a second time interval and a feature signal fusion in an embodiment.

[0132] FIG. 6 is a flowchart of a breathing monitoring joint optimization method in another embodiment.

[0133] FIG. 7 is a flowchart of a breathing monitoring power consumption optimization method in an embodiment.

[0134] FIG. 8 is a flowchart of a breathing monitoring power consumption optimization method in another embodiment.

[0135] FIG. 9 is a waveform diagram of a first time interval and a feature signal fusion in an embodiment.

[0136] FIG. 10 is a flowchart of a breathing monitoring joint optimization method in another embodiment.

[0137] FIG. 11 is a flowchart of a respiratory monitoring joint optimization method in an embodiment.

[0138] FIG. 12 is a flowchart of a respiratory monitoring joint optimization method in another embodiment.

[0139] FIG. 13 is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0140] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0141] In an exemplary embodiment, as shown in FIG. 1, a respiratory monitoring precision optimization method is provided, which is described by taking the case of applying the method to a computer device, including the following steps 102 to 108. The computer device can be a medical respiratory monitoring device, which can be a respiratory monitor, a tongue muscle stimulator, or a hypoglossal nerve stimulator.

[0142] It should be noted that the respiratory monitoring device can determine the sleep respiratory quality according to the index values of the physiological indexes, and determine whether to stimulate the tongue muscle or the hypoglossal nerve to achieve the purpose of intervening the sleep breathing. For example, if the sleep respiratory quality is determined to be low, the tongue muscle or the hypoglossal nerve is stimulated to cause the movement of the tongue, thereby relieving the upper airway obstruction and improving the respiratory quality. Wherein:

[0143] Step 102, obtaining at least two physiological indexes respectively corresponding to a first index value; each first index value is estimated based on a first PPG signal obtained by a sampling controller controlling a PPG sensor to sample.

[0144] The physiological indexes include but are not limited to respiration, heart rate, and blood oxygen.

[0145] The PPG (Photo Plethysmo Graphy) sensor is attached to the chest of the monitored object. The PPG sensor is a sensor that measures blood volume changes using optical technology, which can collect PPG signals for detecting the strength change of reflected light of blood, thereby obtaining heart rate values, blood oxygen content, and respiratory characteristic values corresponding to heart rate, blood oxygen, and respiration, respectively.

[0146] Optionally, the first PPG signal is a correspondence between time instants and received reflected light, wherein a degree of the received reflected light is related to a content of blood oxygen in blood.

[0147] Optionally, when the first PPG signal is obtained by controlling the PPG sensor to sample, the sampling controller is controlled by a sampling parameter, specifically, the sampling controller controls the PPG sensor to sample at a sampling time instant, at a sampling frequency and at an emission intensity. The sampling parameter includes but is not limited to the sampling time instant, the sampling frequency and the emission intensity.

[0148] At step 104, a predicted value of each physiological index in a next breathing cycle is predicted based on each first index value.

[0149] The next breathing cycle is obtained by dividing the time series data of the first PPG signal in units of breathing cycles. The time series data refers to a data sequence arranged in time order, and the time order is a sampling time order.

[0150] Further, each first index value includes a first heart rate value, a first blood oxygen content and a first breathing characteristic value, and the first breathing characteristic value includes a first breathing signal characteristic value and a first time interval of relative static breathing movement.

[0151] The first time interval refers to a sub-time interval in a time interval corresponding to the current breathing cycle. For a breathing cycle, as shown in FIG. 2, during sleep breathing, the reciprocal movement of the chest causes the reflecting surface of the PPG signal in the detection node attached to the chest surface of the monitored object to change, resulting in a correlation between the reflected signal intensity and the breathing movement as shown in FIG. 2.

[0152] Specifically, when the monitored object performs exhalation movement and inhalation movement, i.e. breathing movement, the PPG reflected signal intensity changes periodically, i.e. from the exhalation completion area to the inhalation saturation area, or from the inhalation saturation area to the exhalation completion area, which is one breathing cycle.

[0153] It can be understood that the current breathing cycle is the breathing cycle in which the current breathing movement of the monitored object is performed, which is reflected in the PPG signal, i.e. the breathing cycle in which the latest measured PPG reflected signal intensity is located. Accordingly, the next breathing cycle is the breathing cycle in which the breathing movement of the monitored object has not yet been performed.

[0154] For predicting a predicted value of each physiological index in a next breathing cycle based on each first index value, the predicted value includes a second breathing signal characteristic value, a second blood oxygen content and a second time interval. The second breathing signal characteristic value, the second blood oxygen content and the second time interval predicted in the next breathing cycle are described as follows:

[0155] The second respiratory signal features include the second cycle, the second amplitude, and the second phase, where the second amplitude A i+1 A can be predicted using the following formula: i+1 =A i +f i (H′ i ,H″ i ,H″′ i ,…)+g i (Spo2′ i ,Spo2″ i ,Spo2″′ i ,…) (one)

[0156] Among them, A i f is the first amplitude in the first respiratory signal characteristic value. i (·) and g i (·) represents an empirical formula, where i is the current respiratory cycle number, and H i The heart rate for the current cycle, specifically, H′ i ,H″ i ,H″′ i ... represent multiple first heart rate values ​​for the current respiratory cycle, Spo2′ i ,Spo2″ i ,Spo2″′ i ... represent the first blood oxygen content of the current respiratory cycle.

[0157] Accordingly, the second period and the second phase can be determined by a formula similar to Formula 1, which will not be elaborated here.

[0158] Second blood oxygen saturation Spo2 i+1 It can be predicted using the following formula:

[0159]

[0160] Where i is the sequence number of the current respiratory cycle; A is the second amplitude; and T is the second cycle. This is the second phase; For A, T and The vector formed; Spo2 i It is a vector composed of multiple first blood oxygen contents.

[0161] The second time interval can be defined by the second starting time t′. o_stop Second end time t′ i_start Determined, where the second starting time t′ o_stop The value t′ can be predicted using Formula 3. o_stop =t o_stop +f o(H′ i ,H″ i ,H″′ i ,…) (three)

[0162] Among them, t o_stop f represents the first start time of the first time interval. o (·) is an empirical formula, H′ i ,H″ i ,H″′ i ... represent multiple first heart rate values ​​for the current respiratory cycle.

[0163] The second ending time can be predicted using Formula 4: t′ i_start =t i_start +f i (H′ i ,H″ i ,H″′ i ,…) (Four)

[0164] Among them, t i_start f is the first end time of the first time interval; i (·) is an empirical formula, H′ i ,H″ i ,H″′ i ... represent multiple first heart rate values ​​for the current respiratory cycle.

[0165] Step 106: Based on each first index value and each predicted value, determine the target sampling parameters for the next respiratory cycle when the estimation accuracy is highest; the estimation accuracy is the accuracy when the second index values ​​for the next respiratory cycle are estimated based on the second PPG signal corresponding to the next respiratory cycle; the target sampling parameters include the sampling frequency and the emission intensity of the PPG sensor's emission unit.

[0166] The PPG sensor includes a transmitting unit and a receiving unit. The transmitting unit is used to emit light, and the receiving unit is used to receive the reflected light corresponding to the emitted light.

[0167] Furthermore, the target sampling parameters also include multiple sampling times. When determining the target sampling parameters for the next respiratory cycle with the highest estimation accuracy based on each first index value and each predicted value, the lower bound of error estimation is first used as the optimization target. Based on the end time of the first time interval, the sampling value at that end time, and the first blood oxygen content, each sampling time is determined. Then, based on each first index value and each predicted value, the sampling frequency and the emission intensity of the PPG sensor's emission unit for the next respiratory cycle with the highest estimation accuracy corresponding to each sampling time are determined respectively.

[0168] The error estimation lower bound refers to the minimum difference between the calculation result and the true value. In this embodiment, the calculation result refers to each sampling time.

[0169] The prediction value includes at least the second time interval of the respiratory motion in the next respiratory cycle when it is relatively static, and the sampling time includes at least the first time corresponding to the maximum signal intensity in the next respiratory cycle and at least one second time in the second time interval. That is, the sampling time includes at least the time in the second time interval corresponding to the heart rate or blood oxygen in the next respiratory cycle, and the time in the time interval corresponding to the respiration in the next respiratory cycle, so as to consider the signal characteristics of the PPG signal in different characteristic time intervals, and improve the accuracy of each sampling time.

[0170] It should be noted that the target sampling parameter is determined at the end of the current respiratory cycle, that is, at the end of the first time interval.

[0171] For example, the sampling time is determined, that is, the time sequence signal of the next respiratory cycle is predicted, and when the sampling time is determined, the following formula five can be used to determine it:

[0172] Where i is the serial number of the respiratory cycle, p i is the prediction probability, l(·) is the probability calculation formula, s i is the sampling time, μ i is the mean of the PPG signal intensity in the prediction interval, and σ i is the signal distribution of the time sequence signal.

[0173] Correspondingly, the first sampling time t i1 after the end of the first time interval can be calculated by the formula p i1 . It should be noted that the distribution parameters in the formula, including μ i1 and σ i1 , are affected by the oxygen content, so before predicting the sampling time t i1 , the distribution parameters need to be updated by the following formula six and formula seven:

[0174] μ i1 = μ i0 +z1(Spo2) (six)

[0175] σ i1 = σ i0 +z2(Spo2) (seven)

[0176] Where μ i0 is the mean value at the end of the first time interval, that is, the sampling time t i1the mean value of the previous sampling time of t i0 the signal distribution of the timing signal at the end of the first time interval; z1(Spo2) and z2(Spo2) are the normal distribution of blood oxygen content.

[0177] As shown in FIG. 3, t i0 is the end time of the first time interval, and each sampling time can include t i1 -t i5 The solid line in the prediction interval is drawn through the mean value corresponding to each sampling time.

[0178] In determining the sampling frequency and emission intensity corresponding to each sampling time, the following formula eight can be used: maxz=S(t,f,q,…) (eight)

[0179] Wherein, maxz is the maximum estimation accuracy, S(·) is the estimation accuracy calculation function, t is the sampling time, f is the sampling frequency, and q is the emission intensity. It should be noted that each first index value and each prediction value is a known quantity, which is not shown in formula eight.

[0180] Exemplarily, the target sampling parameters corresponding to each sampling time can be predicted in an alternating manner of prediction and sampling. For example, t i1 After the target sampling parameters corresponding to each sampling time are predicted, the PPG sensor is controlled by the sampling controller to sample at the target sampling parameters, and after sampling, the range of the prediction interval is updated according to the sampling value, so that the estimation accuracy is higher.

[0181] Step 108, based on the target sampling parameters, the sampling parameters of the sampling controller are adjusted, and in the next breathing cycle sampling, the third index value of each physiological index is obtained based on the adjusted sampling parameters, and the prediction function parameters are updated based on the third index value; the prediction function parameters include the calculation parameters corresponding to blood oxygen, heart rate and respiration.

[0182] Wherein, the calculation parameters are the parameters in the prediction function (i.e. the above formulas one to seven) used in the above prediction process.

[0183] Optionally, the original sampling parameters in the sampling controller are updated and adjusted by the target sampling parameters, and then the sensor is controlled to sample by the adjusted sampling parameters to obtain the PPG signal, and then the third index value of each physiological index is obtained by the PPG signal, and the prediction function parameters are updated accordingly. That is, the prediction function parameters are updated in real time based on the corresponding index value after sampling by the multi-feature network control, so as to improve the accuracy of the subsequent index value.

[0184] The method predicts the physiological indicators in the next breathing cycle by the index value, and then determines the target sampling parameter of the next breathing cycle with the highest estimation accuracy based on the first index value and the predicted value, adjusts the sampling controller based on the target sampling parameter, updates the sampling parameter of each physiological indicator, and obtains the third index value of each physiological indicator based on the updated sampling parameter after sampling in the next breathing cycle, and updates the prediction function parameter based on the third index value. The prediction function parameter includes the calculation parameters corresponding to blood oxygen, heart rate and respiration. Therefore, before sampling the sensor signal in the next breathing cycle, the predicted value of each physiological indicator in the next breathing cycle is predicted, and the target sampling parameter is determined based on at least the predicted value. That is, the sampling parameter used in subsequent sampling is not only determined by the current estimated index value, but also combined with the predicted value, so that the target sampling parameter is more suitable for sampling in the next breathing cycle, thereby improving the optimization effect. On this basis, the prediction function parameter is updated according to the obtained third index value after sampling in the next breathing cycle, so as to realize multi-feature network control and improve the estimation accuracy of the estimated index value.

[0185] In one example embodiment, the method further comprises steps 202-208. Wherein:

[0186] Step 202, obtaining a second PPG signal sampled by the PPG sensor in the next breathing cycle, and a second time interval of the predicted respiratory movement in the relative static state.

[0187] Step 204, determining a third time interval in the next breathing cycle except the second time interval.

[0188] Step 206, filtering the second PPG signal in the second time interval and the third time interval based on the prior parameters corresponding to each physiological indicator.

[0189] Step 208, estimating the second index value of each physiological indicator based on the filtered second PPG signal.

[0190] Wherein, each physiological indicator has different characteristic signals, that is, different characteristics on the PPG signal, and the characteristics correspond to the prior parameters of the characteristic signals.

[0191] Exemplarily, in order to avoid loss of information, the signal characteristics corresponding to the second time interval and the third time interval are filtered by the prior parameters corresponding to the corresponding physiological indicators, instead of filtering the complete second PPG signal multiple times by the prior parameters corresponding to all physiological indicators.

[0192] For example, the at least two physiological indexes are respiration and blood oxygen respectively, when filtering the second PPG signal, the second PPG signal is filtered by the prior parameter corresponding to respiration in the second time interval corresponding to respiration, and the second PPG signal is filtered by the prior parameter corresponding to blood oxygen in the third time interval corresponding to blood oxygen. For example, there are time interval 1 and time interval 2 on the second PPG signal, wherein the time interval 1 corresponds to respiration, and the time interval 2 corresponds to blood oxygen, the prior parameter corresponding to respiration is used to filter the signal characteristics of the second PPG signal corresponding to the time interval 1, and the prior parameter corresponding to blood oxygen is used to filter the signal characteristics of the second PPG signal corresponding to the time interval 2.

[0193] It should be noted that after filtering the second PPG signal, the characteristic PPG signal corresponding to each physiological index can be extracted therefrom.

[0194] It should be noted that when calculating the index values corresponding to heart rate and blood oxygen, the signal characteristics corresponding to the time interval of the respiratory movement in the relative static state in the second PPG signal are generally used for calculation. The accuracy of the index values calculated by the signal characteristics is relatively high.

[0195] Therefore, the third time interval corresponding to heart rate and blood oxygen can be the same, and the third time interval is a time interval in a respiratory cycle except the second time interval corresponding to respiration. As shown in FIG. 4, FIG. 4 is a waveform of the characteristic PPG signal corresponding to respiration, wherein the first time interval corresponding to respiration includes t o_start -t o_stop and t i_start -t i_stop , and the first time interval corresponding to heart rate and blood oxygen is t o_stop -t i_start .

[0196] In an embodiment, if there is a same part in the time interval corresponding to at least two physiological indexes respectively, the same part of the time interval can be divided into intervals in a certain proportion, so that the corresponding physiological indexes correspond to different divided intervals respectively. For example, there is a same part time interval 3 in the time interval 1 corresponding to respiration and the time interval 2 corresponding to blood oxygen, the time interval 3 is divided into two parts in an equal division manner, and the two parts correspond to respiration and blood oxygen respectively.

[0197] In another embodiment, for filtering the second PPG signal based on the prior parameters corresponding to each physiological index respectively in the second time interval and the third time interval, it specifically includes:

[0198] If there are two target physiological indicators corresponding to the second time interval in at least two physiological indicators, the correlation between each sampling data corresponding to the second time interval in the second PPG signal and each target physiological indicator is determined respectively; and each sampling data is filtered based on the prior parameter corresponding to the target physiological indicator with the highest correlation.

[0199] It should be noted that the second PPG signal is actually composed of a plurality of sampling data, and the waveform thereof is formed by connecting the points corresponding to the sampling values of the sampling data.

[0200] The correlation refers to the matching degree between the sampling data and the prior parameter corresponding to each physiological indicator.

[0201] Exemplarily, there is different correlation between each sampling data and different physiological indicators, and when each sampling data is filtered based on the prior parameter corresponding to the physiological indicator with the highest correlation, the information of the sampling data can be retained more, thereby improving the accuracy of the estimated first indicator value.

[0202] In the embodiment, when the indicator value is determined by the PPG signal, the characteristics corresponding to different physiological indicators are considered to be related to each other, and the overall PPG signal is not filtered respectively, but the segmented filtering is realized by the corresponding time interval. That is, the PPG signal corresponding to each time interval is only filtered by the physiological indicator corresponding thereto, rather than all physiological indicators, thereby avoiding the loss of information in the PPG signal and improving the accuracy of each indicator value.

[0203] On the basis of the above embodiments, each second indicator value includes a second heart rate value, a second blood oxygen content, and a second respiratory characteristic value, and the second indicator value of each physiological indicator estimated based on the filtered second PPG signal includes the following steps 302 to 308. Wherein:

[0204] Step 302, the respiratory characteristic PPG signal corresponding to the respiration, the blood oxygen characteristic PPG signal corresponding to the blood oxygen, and the heart rate characteristic PPG signal corresponding to the heart rate are extracted from the filtered second PPG signal respectively.

[0205] Each characteristic PPG signal is estimated by the correlation calculation between the filtered second PPG signal and the corresponding characteristic signal, that is, the characteristic signal corresponding to the respiration can estimate the respiratory characteristic PPG signal, the characteristic signal corresponding to the blood oxygen can estimate the blood oxygen characteristic PPG signal, and the characteristic signal corresponding to the heart rate can estimate the heart rate characteristic PPG signal.

[0206] Step 304, the second respiratory characteristic value is determined based on the respiratory characteristic PPG signal; the second respiratory characteristic value includes a second respiratory signal characteristic value and a fourth time interval of the respiratory movement in the relative static state.

[0207] The second respiratory signal feature value comprises a second period, a second amplitude and a second phase.

[0208] Optionally, the second respiratory feature value can be directly identified from the waveform corresponding to the respiratory feature PPG signal.

[0209] As shown in FIG. 4, which is a waveform diagram corresponding to the respiratory feature PPG signal, for the Ti period, the exhalation interval t o_start -t o_stop , the inhalation interval t i_start -t i_stop , and t o_stop -t i_start are the fourth time interval in the relative static state.

[0210] Step 306, determining the second blood oxygen content based on the fourth time interval and the blood oxygen feature PPG signal.

[0211] As shown in FIG. 5, the fourth time interval is fused with the blood oxygen feature PPG signal, that is, the signal feature of the blood oxygen feature PPG signal corresponding to the fourth time interval, i.e., t o_stop -t i_start , is determined in the blood oxygen feature PPG signal. Then the second blood oxygen content is determined through the signal feature.

[0212] Further, when the second blood oxygen content is determined through the signal feature, the sampling data is obtained from the signal feature; then the second blood oxygen content is calculated through the sampling data.

[0213] The calculation process is: determining the first alternating component and the first direct current component of the red light signal, and the second alternating component and the second direct current component of the infrared signal in the sampling data; obtaining the second blood oxygen content based on the first alternating component, the first direct current component, the second alternating component and the second direct current component.

[0214] Referring to the following Formula Nine, the second blood oxygen content Spo2 can be calculated through the Formula Nine:

[0215] Wherein, Ac red is the first alternating component of the red light signal, Dc red is the first direct current component of the red light signal, Ac ir is the second alternating component of the infrared signal, and Dc ir is the second direct current component of the infrared signal.

[0216] Step 308, determining the second heart rate value based on the fourth time interval and the heart rate feature PPG signal.

[0217] Similarly, as shown in FIG. 5, the fourth time interval is fused with the heart rate feature PPG signal, i.e. the signal feature of the heart rate feature PPG signal corresponding to the fourth time interval is determined in the heart rate feature PPG signal, i.e. t o_stop -t i_start the corresponding signal feature. Then the second heart rate value is determined through the signal feature.

[0218] In an exemplary embodiment, as shown in FIG. 6, the overall flow of the method can include steps 1-16. Among them:

[0219] Step 1: Based on the sampling controller, the PPG sensor (including PPG emitting unit and PPG receiving unit) is controlled to sample at sampling time, sampling frequency and emission intensity to obtain a PPG signal, and the PPG signal is input into a PPG feature extractor. Wherein, the PPG sensor as a respiration monitoring node can be attached to the chest of the monitored object, or placed in the chest cavity.

[0220] Step 2: The PPG signal is processed in advance, filtering, etc.

[0221] Step 3: Extract the respiratory feature PPG.

[0222] Step 4: Extract the blood oxygen feature PPG.

[0223] Step 5: Extract the heart rate feature PPG.

[0224] Step 6: Based on the respiratory feature PPG, the respiratory motion is detected, and the first respiratory signal feature value and the first time interval of the respiratory motion in the relative static state are detected.

[0225] Step 7: Based on the blood oxygen feature PPG and the first time interval obtained in subsequent step 9, the blood oxygen content is calculated.

[0226] Step 8: Based on the heart rate feature PPG and the first time interval obtained in subsequent step 9, the heart rate value is calculated.

[0227] Step 9: The first time interval is input into the blood oxygen content calculation module and the heart rate calculation module, so that the blood oxygen content calculation module calculates the first blood oxygen content, and the heart rate calculation module calculates the first heart rate value.

[0228] Step 10: inputting the first time interval into the respiration prediction module; inputting the first heart rate value obtained in step 12 into the respiration prediction module to predict the respiration of the next breathing cycle, and obtaining the second time interval of the respiration movement in the relative static state; and inputting the first blood oxygen content obtained in step 11 and the first heart rate value obtained in step 12 into the respiration prediction module to predict the second respiration signal characteristic value of the next breathing cycle.

[0229] Step 11: inputting the first blood oxygen content into the blood oxygen prediction module to predict the second blood oxygen content of the next breathing cycle based on the first blood oxygen content and the second respiration signal characteristic value obtained in step 13.

[0230] Step 12: inputting the first heart rate value into the respiration prediction module.

[0231] Step 13: inputting the second respiration signal characteristic value into the blood oxygen prediction module by the respiration prediction module.

[0232] Step 14: inputting the second respiration signal characteristic value and the second time interval predicted by the respiration prediction module into the estimation accuracy optimizer. The estimation accuracy optimizer optimizes the estimation accuracy based on the index value and the predicted value to obtain the target sampling parameter. The estimation accuracy optimizer also obtains the predicted value of other physiological indexes of the next breathing cycle.

[0233] Step 15: inputting the predicted value obtained in the prediction process, such as the second time interval, into the PPG feature extractor.

[0234] Step 16: inputting the target sampling parameter into the sampling controller to adjust the sampling parameter by the sampling controller, and obtaining the third index value of each physiological index based on the adjusted sampling parameter when sampling in the next breathing cycle, and updating the prediction function parameter based on the third index value; the prediction function parameter includes the calculation parameter corresponding to blood oxygen, heart rate and respiration.

[0235] In the related art, when monitoring physiological indexes, two or more sensors can be used for monitoring. For example, an IMU (Inertial Measurement Unit) sensor and a PPG sensor are used for monitoring.

[0236] However, the sensors consume the power of the battery when in use. At present, the method of monitoring physiological indexes by using two or more sensors aggravates the consumption of the power of the respiratory monitoring device, resulting in low endurance of the respiratory monitoring device. The respiratory monitoring nodes of the respiratory monitoring device include wearable and implantable.

[0237] In one exemplary embodiment, as shown in FIG. 7, a respiratory monitoring power consumption optimization method is provided, which is applied to a computer device for example, and includes the following steps 702 to 706. The computer device can be a respiratory monitoring device for medical use. The respiratory monitoring device can be a respiratory monitor, a tongue muscle stimulator, or a hypoglossal nerve stimulator.

[0238] In step 702, first index values respectively corresponding to a current respiratory cycle and second index values respectively corresponding to a previous respiratory cycle of at least two physiological indexes are obtained, and a power consumption characteristic coefficient of a light source of a PPG sensor is obtained. Each first index value is estimated based on a first PPG signal obtained by sampling the PPG sensor under the control of a sampling controller. The light source includes a red light source and an infrared light source.

[0239] The physiological indexes include but are not limited to respiration, heart rate, and blood oxygen. The index values are values corresponding to the physiological indexes, such as heart rate values and blood oxygen content.

[0240] The previous respiratory cycle is relative to the current respiratory cycle. As shown in FIG. 2, when the monitored object performs an exhalation movement and an inhalation movement, i.e., a respiratory movement, the PPG reflection signal intensity changes periodically, i.e., from the exhalation completion area to the inhalation saturation area, or from the inhalation saturation area to the exhalation completion area, which is one respiratory cycle.

[0241] It can be understood that the current respiratory cycle is the respiratory cycle in which the current respiratory movement of the monitored object is performed, which is reflected in that the first PPG signal is the respiratory cycle in which the latest measured PPG reflection signal intensity is located. Accordingly, the previous respiratory cycle is the respiratory cycle in which the respiratory movement of the monitored object has been performed.

[0242] The power consumption characteristic coefficient includes a power consumption characteristic coefficient γ1 of the red light source and a power consumption characteristic coefficient γ2 of the infrared light source. It can be understood that the power consumption characteristic coefficient represents the power consumption of the light source.

[0243] Exemplarily, the sampling controller controls the PPG sensor with sampling parameters. The sampling parameters include but are not limited to a sampling time, a sampling period, and a sampling intensity.

[0244] Optionally, when the first PPG signal is obtained by sampling the PPG sensor under the control of the sampling controller, the sampling controller controls the PPG sensor to sample at a sampling time, with a sampling frequency and a sampling intensity.

[0245] In step 704, predicted values of each physiological index in a next respiratory cycle are predicted based on each first index value.

[0246] The next respiratory cycle is the respiratory cycle in which the respiratory movement of the monitored object has not been performed.

[0247] wherein each first index value comprises a first heart rate value, a first blood oxygen content, and a first respiration characteristic value, the first respiration characteristic value comprising a first respiration signal characteristic value and a first time interval of the respiration movement at a relative static state.

[0248] For predicting each physiological index in the next respiration cycle based on each first index value, a predicted value is obtained, which comprises a third respiration signal characteristic value, a third blood oxygen content, and a third time interval. The third respiration signal characteristic value, the third blood oxygen content, and the third time interval in the next respiration cycle are described as follows respectively:

[0249] The third respiration signal characteristic value comprises a third period, a third amplitude, and a third phase, wherein the third amplitude A i+1 may be predicted by the following Formula One: i+1 = A i +f i (H′ i ,H″ i ,H″′ i ,…)+g i (Spo2′ i ,Spo2″ i ,Spo2″′ i ,…) (One)

[0250] wherein A i is the first amplitude in the first respiration signal characteristic value, f i (·) and g i (·) are empirical formulas, i is the serial number of the current respiration cycle, H′ i ,H″ i ,H″′ i ,… are a plurality of first heart rate values in the current respiration cycle, Spo2′ i ,Spo2″ i ,Spo2″′ i ,… are a plurality of first blood oxygen contents in the current respiration cycle.

[0251] Correspondingly, the third period and the third phase can be determined by a formula similar to Formula One, which is not described herein again.

[0252] The third blood oxygen content Spo2 i+1 may be predicted by the following Formula Two:

[0253] wherein i is the serial number of the current respiration cycle; A is the third amplitude, T is the third period, is the third phase; is a function of A, T, and Spo2 i is a vector composed of a plurality of first blood oxygen contents.

[0254] Further, when predicting the third time interval of the respiratory movement in the relatively static state of the next breathing cycle based on the first time interval and the first heart rate value, the third time interval is determined by:

[0255] The third time interval can be determined by a third start time t' o_stop and a third end time t' i_start , wherein the third start time t' o_stop can be predicted by formula three: t' o_stop = t o_stop + f o (H' i , H" i , H" i ,...) (Three)

[0256] Wherein t o_stop is the first start time of the first time interval; f o (·) is an empirical formula, H' i , H" i , H" i , … are a plurality of first heart rate values of the current breathing cycle.

[0257] The third end time t' i_start can be predicted by formula four: t' i_start = t i_start + f i (H' i , H" i , H" i ,...) (Four)

[0258] Wherein t i_start is the first end time of the first time interval; f i (·) is an empirical formula, H' i , H" i , H" i , … are a plurality of first heart rate values of the current breathing cycle.

[0259] Wherein f o (H' i , H" i , H" i , …) is the first gain, and f i (H' i , H" i , H" i , …) is the second gain.

[0260] That is, in determining the third time interval, a first gain of a first start time and a second gain of a first end time of the first time interval are determined based on the first heart rate value respectively; a third start time of the third time interval when the respiratory movement of the next breathing cycle is relatively static is determined based on the first start time and the first gain; a third end time of the third time interval when the respiratory movement of the next breathing cycle is relatively static is determined based on the first end time and the second gain.

[0261] In step 706, based on the power consumption characteristic coefficient, the second index value and the predicted value, the target sampling parameter of the PPG sensor corresponding to the next breathing cycle when the power consumption is the lowest is determined.

[0262] In an embodiment, the target sampling parameter of the PPG sensor corresponding to the next breathing cycle when the power consumption is the lowest can be determined by the following formula five based on the power consumption characteristic parameter and the first PPG signal:

[0263] Wherein, w is the power consumption of the PPG sensor, T is the third period, A is the third amplitude, is the third phase; γ1 is the power consumption characteristic coefficient of the red light source, γ2 is the power consumption characteristic coefficient of the infrared light source; t is the sampling time, σ is the sampling intensity, and P is the sampling period.

[0264] Exemplarily, in formula five, T, A, γ1 and γ2 are known quantities, t, σ and P are unknown quantities, and when determining the target sampling parameter of the PPG sensor corresponding to the next breathing cycle when the power consumption is the lowest, t, σ and P when the power consumption w is the minimum are actually calculated.

[0265] It should be noted that each second index value includes a second heart rate value, a second blood oxygen content and a second respiratory characteristic value, the second respiratory characteristic value includes a second respiratory signal characteristic value and a second time interval when the respiratory movement is relatively static, and the second respiratory signal characteristic value includes a second period T, a second amplitude A and a second phase

[0266] It should be noted that the first PPG signal includes sampling data of the current breathing cycle and sampling data of the previous breathing cycle, that is, through the first PPG signal, the index values of each physiological index in the current breathing cycle and any previous breathing cycle can be determined, including the second index value of the previous breathing cycle.

[0267] Wherein, the time sequence data of the first PPG signal is divided in units of breathing cycles, that is, the first PPG signal records the PPG reflection signal intensity of the current breathing cycle and any previous breathing cycle.

[0268] In another embodiment, the target sampling parameter is determined similar to that by Equation 5. Also, t, σ and P at which the power consumption is the lowest can be calculated, with the difference being that the known quantities further include the predicted value, and the calculation process will not be repeated here.

[0269] At step 708, the sampling controller is adjusted in sampling parameter based on the target sampling parameter, and at the next breathing cycle sampling, the third index value of each physiological index is obtained based on the adjusted sampling parameter, and the prediction function parameter is updated based on the third index value; the prediction function parameter includes the calculation parameter corresponding to blood oxygen, heart rate and respiration.

[0270] The calculation parameter is a parameter in the prediction function (i.e. Equations 1 to 4) used in the above prediction process.

[0271] Optionally, the original sampling parameter in the sampling controller is adjusted in parameter update based on the target sampling parameter, and then the PPG signal is obtained by controlling the sensor to sample based on the adjusted sampling parameter, and then the third index value of each physiological index is obtained based on the PPG signal, and the prediction function parameter is updated based on the third index value. That is, the prediction function parameter is updated in multi-feature network control in real time based on the corresponding index value after sampling, so as to improve the accuracy of the subsequent index value.

[0272] The above method determines the index value of at least two physiological indexes only by the PPG sensor, and on this basis, the predicted value of the next breathing cycle is obtained by predicting based on the first index value of the current breathing cycle, and the target sampling parameter of the PPG sensor corresponding to the next breathing cycle at which the power consumption is the lowest is determined based on the predicted value, the second index value of the previous breathing cycle and the power consumption characteristic coefficient of the light source, so as to optimize the sampling controller. Thus, the need for two sensors in breathing monitoring is avoided, and the power consumption of the PPG sensor is also reduced based on the use of one PPG sensor, thereby reducing the consumption of electric energy. In addition, after sampling in the next breathing cycle, the prediction function parameter is updated based on the obtained third index value, the accuracy of the predicted value is improved, the power consumption calculation is more accurate, the power consumption is further reduced, and the endurance of the breathing monitoring device is improved.

[0273] In one exemplary embodiment, as shown in FIG. 8, the above step 702 includes steps 7022-7028. Wherein:

[0274] At step 7022, the first PPG signal obtained by the PPG sensor in the current breathing cycle is obtained, and the fourth time interval of the respiratory movement in the current breathing cycle at the relative static state is predicted in the previous breathing cycle.

[0275] At step 7024, the fifth time interval in the current breathing cycle except the fourth time interval is determined.

[0276] At step 7026, the first PPG signal is filtered in the fourth time interval and the fifth time interval respectively based on the prior parameters corresponding to the physiological indexes.

[0277] At step 7028, the first index values corresponding to the physiological indexes respectively are estimated based on the filtered first PPG signal.

[0278] The physiological indexes have different characteristic signals, i.e., different characteristics on the PPG signal, which correspond to the prior parameters of the characteristic signals.

[0279] For example, to avoid loss of information, the signal characteristics corresponding to the fourth time interval and the fifth time interval are filtered by the prior parameters corresponding to the physiological indexes, instead of filtering the complete first PPG signal by the prior parameters corresponding to all the physiological indexes.

[0280] For example, the at least two physiological indexes are respiration and blood oxygen respectively, and when the first PPG signal is filtered, the first PPG signal is filtered in the fourth time interval corresponding to respiration by the prior parameters corresponding to respiration, and the first PPG signal is filtered in the fifth time interval corresponding to blood oxygen by the prior parameters corresponding to blood oxygen. For example, the first PPG signal has a time interval 1 and a time interval 2, wherein the time interval 1 corresponds to respiration and the time interval 2 corresponds to blood oxygen, the prior parameters corresponding to respiration are used to filter the signal characteristics of the first PPG signal corresponding to the time interval 1, and the prior parameters corresponding to blood oxygen are used to filter the signal characteristics of the first PPG signal corresponding to the time interval 2.

[0281] It should be noted that after the first PPG signal is filtered, the characteristic PPG signals corresponding to the physiological indexes can be extracted therefrom.

[0282] It should be noted that when the index values corresponding to the heart rate and blood oxygen are calculated, the signal characteristics corresponding to the time interval of the respiratory motion in the first PPG signal in the relatively static state are generally used for calculation. The accuracy of the index values calculated by the signal characteristics is relatively high.

[0283] Therefore, the fifth time interval corresponding to the heart rate and blood oxygen can be the same, and the fifth time interval is a time interval in a respiratory cycle except the fourth time interval corresponding to respiration. As shown in FIG. 4, FIG. 4 is a waveform of the characteristic PPG signal corresponding to respiration, wherein the first time interval corresponding to respiration includes t o_start -t o_stop and t i_start -t i_stop , and the first time interval corresponding to the heart rate and blood oxygen is t o_stop -ti_start .

[0284] In an embodiment, if there is a same part in the time interval corresponding to each of the at least two physiological indicators, the time interval of the same part can be divided into intervals in a certain proportion, so that the corresponding physiological indicators correspond to different divided intervals respectively. For example, there is a same part time interval 3 in the time interval 1 corresponding to the respiration and the time interval 2 corresponding to the blood oxygen. The time interval 3 is divided into two equal parts, which correspond to the respiration and the blood oxygen respectively.

[0285] In another embodiment, based on the prior parameters corresponding to each physiological indicator, the first PPG signal is filtered in the fourth time interval and the fifth time interval respectively, which specifically includes:

[0286] If there are two target physiological indicators corresponding to the fourth time interval in the at least two physiological indicators, the correlation degree between each sampling data corresponding to the fourth time interval in the first PPG signal and each target physiological indicator is determined respectively; each sampling data is filtered based on the prior parameter corresponding to the target physiological indicator with the highest correlation degree.

[0287] It should be noted that the first PPG signal is actually composed of a plurality of sampling data, and the waveform is connected by the points corresponding to the sampling values of the sampling data.

[0288] Among them, the correlation degree refers to the matching degree between the sampling data and the prior parameters corresponding to each physiological indicator.

[0289] Exemplarily, there is a different correlation degree between each sampling data and different physiological indicators. When each sampling data is filtered by the prior parameter corresponding to the physiological indicator with the highest correlation degree, the information of the sampling data can be retained more, thereby improving the accuracy of the subsequent estimated first indicator value.

[0290] In this embodiment, when determining the indicator value through the PPG signal, the characteristics corresponding to different physiological indicators are considered to be related to each other, and the overall PPG signal is not filtered by various filters respectively, but is segmented and filtered by the corresponding time interval. That is, the PPG signal corresponding to each time interval is only filtered by the physiological indicator corresponding to the physiological indicator, rather than all physiological indicators, thereby avoiding the loss of information in the PPG signal and improving the accuracy of each indicator value.

[0291] Further, each first indicator value includes a first respiration characteristic value, a first blood oxygen content and a first heart rate value, and the above-mentioned estimation of each physiological indicator corresponding to the first indicator value based on the filtered first PPG signal includes steps 902 to 908. Among them:

[0292] Step 902, respectively extracting a respiration feature PPG signal corresponding to respiration, a blood oxygen feature PPG signal corresponding to blood oxygen, and a heart rate feature PPG signal corresponding to heart rate from the filtered first PPG signal.

[0293] Wherein, each feature PPG signal is estimated by correlation degree calculation of the first PPG signal and the corresponding feature signal, that is, the respiration feature signal corresponding to respiration can estimate the respiration feature PPG signal, the blood oxygen feature signal corresponding to blood oxygen can estimate the blood oxygen feature PPG signal, and the heart rate feature signal corresponding to heart rate can estimate the heart rate feature PPG signal.

[0294] Step 904, determining a first respiration feature value based on the respiration feature PPG signal; the first respiration feature value includes a first respiration signal feature value and a first time interval of relative static state of respiration movement.

[0295] Wherein, the first respiration signal feature value includes a first period, a first amplitude and a first phase.

[0296] Optionally, the first respiration feature value can be directly identified from the waveform corresponding to the respiration feature PPG signal.

[0297] As shown in FIG. 4, for Ti period, the exhalation interval t o_start -t o_stop , the inhalation interval t i_start -t i_stop , and t o_stop -t i_start is the first time interval of relative static state.

[0298] Step 906, determining a first blood oxygen content based on the first time interval and the blood oxygen feature PPG signal.

[0299] As shown in FIG. 9, fusing the first time interval with the blood oxygen feature PPG signal, that is, determining the first blood oxygen content based on the fused blood oxygen feature PPG signal.

[0300] In an embodiment, the process of determining the first blood oxygen content based on the fused blood oxygen feature PPG signal is: determining the signal feature of the blood oxygen feature PPG signal corresponding to the first time interval in the blood oxygen feature PPG signal, that is, the signal feature corresponding to to_stop-ti_start. Then, determining the first blood oxygen content based on the signal feature.

[0301] Further, when determining the first blood oxygen content based on the signal feature, the sampling data is obtained from the signal feature; then, the first blood oxygen content is calculated through the sampling data.

[0302] The calculation process is: determining the first alternating component and the first direct current component of the red light signal and the second alternating component and the second direct current component of the infrared signal in the sampling data; obtaining the first blood oxygen content based on the first alternating component, the first direct current component, the second alternating component and the second direct current component.

[0303] Referring to the following formula six, the first blood oxygen content Spo2 can be calculated by the formula six:

[0304] Wherein, Ac red is the first alternating component of the red light signal, Dc red is the first direct current component of the red light signal, Ac ir is the second alternating component of the infrared signal, Dc ir is the second direct current component of the infrared signal.

[0305] Step 908, determining the first heart rate value based on the first time interval and the heart rate characteristic PPG signal.

[0306] Similarly, as shown in FIG. 9, the first time interval is fused with the heart rate characteristic PPG signal, that is, the signal feature of the heart rate characteristic PPG signal corresponding to the first time interval in the heart rate characteristic PPG signal can be determined, that is, t o_stop -t i_start The corresponding signal feature. Then the first heart rate value is determined through the signal feature.

[0307] In an exemplary embodiment, after determining the target sampling parameter of the next breathing cycle corresponding to the lowest power consumption of the PPG sensor based on the power consumption characteristic coefficient, the second index value and the predicted value, the method further comprises:

[0308] Updating the prior parameter corresponding to each physiological index based on the target sampling parameter and the predicted value.

[0309] Wherein, the prior parameter includes but is not limited to the index value and the sampling parameter corresponding to the physiological index. Such as the period corresponding to the respiration, the time interval and the sampling intensity of the respiratory movement in the relative static state.

[0310] It should be noted that based on the updated prior parameter, the second PPG signal of the next breathing cycle can be filtered, thereby improving the estimation accuracy of the third index value corresponding to at least two physiological indexes in the next breathing cycle respectively.

[0311] In an exemplary embodiment, as shown in FIG. 10, the overall flow of the method can include steps 1-17. Wherein:

[0312] Step 1: Control the PPG sensor (including a PPG emitting unit and a PPG receiving unit) to sample at a sampling time, a sampling period and a sampling intensity based on a sampling controller to obtain a PPG signal, and input the PPG signal into a PPG feature extractor. The PPG sensor as a respiratory monitoring node can be attached to the chest of a monitored object or placed in the chest cavity of the monitored object.

[0313] Step 2: Perform priori processing and filtering on the PPG signal.

[0314] Step 3: Extract a respiratory feature PPG.

[0315] Step 4: Extract a blood oxygen feature PPG.

[0316] Step 5: Extract a heart rate feature PPG.

[0317] Step 6: Detect respiratory motion based on the respiratory feature PPG to obtain a first respiratory signal feature value and a first time interval of the respiratory motion in a relatively static state.

[0318] Step 7: Calculate blood oxygen content based on the blood oxygen feature PPG and the first time interval obtained in subsequent step 9.

[0319] Step 8: Calculate a heart rate value based on the heart rate feature PPG and the first time interval obtained in subsequent step 9.

[0320] Step 9: Input the first time interval into a blood oxygen content calculation module and a heart rate calculation module, so that the blood oxygen content calculation module calculates a first blood oxygen content and the heart rate calculation module calculates a first heart rate value.

[0321] Step 10: Input the first time interval into a respiratory prediction module; the respiratory prediction module performs respiratory prediction based on the first time interval and a first heart rate value obtained in subsequent step 12 to obtain a third time interval of respiratory motion in a relatively static state in a next respiratory cycle; and the respiratory prediction module predicts a third respiratory signal feature value in the next respiratory cycle based on a first blood oxygen content obtained in step 11, the first heart rate value obtained in subsequent step 12 and the first time interval.

[0322] Step 11: Input the first blood oxygen content into a blood oxygen prediction module, so that the blood oxygen prediction module predicts a third blood oxygen content in the next respiratory cycle based on the first blood oxygen content and a third respiratory signal feature value obtained in subsequent step 13.

[0323] Step 12: Input the first heart rate value into the respiratory prediction module.

[0324] Step 13: The respiratory prediction module inputs the third respiratory signal feature value into the blood oxygen prediction module.

[0325] Step 14: input the third respiratory signal feature value predicted by the respiratory prediction module and the third time interval into the node power consumption optimizer. The node power consumption optimizer performs estimation accuracy optimization based on the index value, the power consumption characteristic coefficient, and the predicted value to obtain a target sampling parameter. The estimation accuracy optimizer also obtains predicted values of other physiological indexes in the next respiratory cycle.

[0326] Step 15: adaptively adjust the target sampling parameter through the power consumption optimization coefficient corresponding to the red light source and the power consumption optimization coefficient corresponding to the infrared light source, respectively.

[0327] Step 16: input the predicted value obtained in the prediction process, such as the third time interval, into the PPG feature extractor.

[0328] Step 17: input the adjusted target sampling parameter into the sampling controller for sampling parameter adjustment. In the next respiratory cycle, based on the adjusted sampling parameter, the third index value of each physiological index is obtained, and the prediction function parameter is updated based on the third index value. The prediction function parameter includes the calculation parameter corresponding to blood oxygen, heart rate, and respiration.

[0329] In an exemplary embodiment, as shown in FIG. 11, a respiratory monitoring joint optimization method is provided. Taking the case of applying the method to a computer device for example, the method includes the following steps 1102 to 1108. The computer device can be a medical respiratory monitoring device. The respiratory monitoring device can be a respiratory monitor, a tongue muscle stimulator, or a hypoglossal nerve stimulator.

[0330] It should be noted that the respiratory monitoring device can determine the sleep respiratory quality according to the index value of each physiological index, and determine whether to stimulate the tongue muscle or the hypoglossal nerve to achieve the purpose of intervening in the sleep respiration. For example, if the sleep respiratory quality is low, the tongue muscle or the hypoglossal nerve is stimulated to cause the movement of the tongue, thereby relieving the upper airway obstruction and improving the respiratory quality. Wherein:

[0331] Step 1102: obtain at least two physiological indexes corresponding to the first index value in the current respiratory cycle; each first index value is estimated based on the first PPG signal obtained by the sampling controller controlling the PPG sensor to sample.

[0332] Wherein, the physiological index includes but is not limited to respiration, heart rate, and blood oxygen. The index value is the value corresponding to the physiological index, such as the heart rate value and the blood oxygen content.

[0333] The PPG reflection signal strength changes periodically when the monitored object performs the exhalation movement and the inhalation movement, i.e., the respiratory movement, as shown in FIG. 2, i.e., from the exhalation completion area to the inhalation saturation area, or from the inhalation saturation area to the exhalation completion area, which is one respiratory cycle.

[0334] It can be understood that the current respiratory cycle is the respiratory cycle in which the monitored object is currently performing the respiratory movement, and the first PPG signal is the respiratory cycle in which the latest measured PPG reflection signal strength is located.

[0335] The PPG sensor is attached to the chest of the monitored object. The PPG sensor is a sensor that measures blood volume changes using optical technology, which can collect PPG signals for detecting the strength changes of reflected light of blood, thereby obtaining heart rate, blood oxygen, and respiratory characteristics corresponding to heart rate values, blood oxygen content, and respiratory characteristic values, respectively.

[0336] The first PPG signal is the corresponding relationship between the sampling time and the received reflected light, wherein the strength of the received reflected light is positively correlated with the amount of blood oxygen cells in the blood.

[0337] For example, the sampling controller controls the PPG sensor with sampling parameters. The sampling parameters include, but are not limited to, sampling time, sampling frequency, and emission intensity of the emission unit of the PPG sensor.

[0338] Optionally, when the first PPG signal is obtained by controlling the PPG sensor to sample by the sampling controller, the sampling controller controls the PPG sensor to sample at the sampling time, with the sampling frequency and the emission intensity.

[0339] Step 1104, based on each first index value, a predicted value of each physiological index in the next respiratory cycle is obtained.

[0340] The next respiratory cycle is obtained by dividing the time sequence data of the first PPG signal by the respiratory cycle. The time sequence data refers to a data sequence arranged in time sequence, and the time sequence is the sampling time sequence.

[0341] Further, the first index values include the first heart rate value, the first blood oxygen content, and the first respiratory characteristic value, and the first respiratory characteristic value includes the first respiratory signal characteristic value and the first time interval when the respiratory movement is relatively static.

[0342] The first time interval refers to a sub-time interval in a time interval corresponding to a current breathing cycle. For a breathing cycle, as shown in FIG. 2, in a sleep breathing process, the reciprocating movement of the chest causes the reflecting surface of the PPG signal in the detection node (i.e., the transmitting unit and the receiving unit of the PPG sensor) attached to the chest surface of the monitored object to change, resulting in a correlation between the reflected signal intensity and the breathing movement as shown in FIG. 2.

[0343] The prediction value of each physiological indicator in the next breathing cycle is obtained based on each first indicator value. The prediction value includes a second respiratory signal characteristic value, a second blood oxygen content, and a second time interval. The second respiratory signal characteristic value, the second blood oxygen content, and the second time interval in the next breathing cycle are described as follows:

[0344] The second respiratory signal characteristic value includes a second period, a second amplitude, and a second phase, wherein the second amplitude A i+1 The second respiratory signal characteristic value can be predicted by the following formula (I): i+1 i i (H′ i ,H″ i ,H″′ i ,…)+g i (Spo2′ i ,Spo2″ i ,Spo2″′ i ,…) (I)

[0345] wherein A i is the first amplitude in the first respiratory signal characteristic value, f i (·) and g i (·) are empirical formulas, i is the serial number of the current breathing cycle, H′ i ,H″ i ,H″′ i ,… are a plurality of first heart rate values of the current breathing cycle, Spo2′ i ,Spo2″ i ,Spo2″′ i ,… are a plurality of first blood oxygen contents of the current breathing cycle.

[0346] Correspondingly, the second period and the second phase can be determined by a formula similar to formula (I), which is not described here.

[0347] The second blood oxygen content Spo2 i+1 can be predicted by the following formula (II):

[0348] wherein i is the serial number of the current breathing cycle; A is the second amplitude, T is the second period, ​​for the second phase; consisting of A, T and Spo2 i is a vector consisting of the plurality of first blood oxygen contents.

[0349] The second time interval can be determined by a second start time t' o_stop and a second end time t' i_start , wherein the second start time t' o_stop may be predicted by formula three: t' o_stop = t o_stop + f o (H' i , H" i , H" i ,...) (three)

[0350] wherein t o_stop is the first start time of the first time interval; f o (·) is an empirical formula, H' i , H" i , H" i ,... are the plurality of first heart rate values of the current respiratory cycle.

[0351] The second end time can be predicted by formula four: t' i_start = t i_start + f i (H' i , H" i , H" i ,...) (four)

[0352] wherein t i_start is the first end time of the first time interval; f i (·) is an empirical formula, H' i , H" i , H" i ,... are the plurality of first heart rate values of the current respiratory cycle.

[0353] In step 1106, based on the plurality of predicted values, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient, and a current characteristic coefficient of a light source of the PPG sensor, a target sampling parameter corresponding to a minimum value obtained in a joint power consumption calculation and error bound calculation is determined; the light source includes a red light source and an infrared light source; and the target sampling parameter is used for the sampling controller to control the PPG sensor to sample in a next respiratory cycle.

[0354] wherein the preset accuracy optimization coefficient and the preset power consumption optimization coefficient are empirical values.

[0355] The current characteristic coefficient refers to the relationship between the current of the red light source and the current of the infrared light source when the red light source and the infrared light source are working. Specifically, the current of the red light source is stronger than the current of the infrared light source. Therefore, the power consumption of the light source can be reduced according to the relationship.

[0356] The target sampling parameter includes but is not limited to a sampling time, a sampling frequency and an emission intensity.

[0357] In an embodiment, for the determination of the target sampling parameter, the power consumption is calculated based on the prediction values, and the error bound is calculated based on the prediction values. The preset power consumption optimization coefficient and the current characteristic coefficient of the light source of the PPG sensor are used as the first coefficient in the power consumption calculation, and the preset accuracy optimization coefficient is used as the second coefficient in the error bound calculation. The joint calculation is performed to determine the target sampling parameter corresponding to the minimum value obtained in the joint calculation. The determination process can be realized by the following formula five:

[0358] In the formula five, ε1 is the preset accuracy optimization coefficient, ε2 is the preset power consumption optimization coefficient, S1(·) is the estimation accuracy calculation function, S2(·) is the power consumption calculation function, t is the sampling time, f is the sampling frequency, and q is the emission intensity.

[0359] In the formula five, ε1, ε2, the prediction values (not shown, in S1(·) and S2(·)) and the current characteristic coefficient (not shown, in S2(·)) are known quantities, and t, f and q are unknown quantities.

[0360] It can be understood that the target sampling parameter is determined when the minimum value is obtained in the joint power consumption calculation and error bound calculation. In fact, t, f and q are calculated when z is the minimum value.

[0361] In step 1108, the sampling parameter of the sampling controller is adjusted based on the target sampling parameter.

[0362] Optionally, the original sampling parameter in the sampling controller is updated by the target sampling parameter.

[0363] Further, in the next breathing cycle sampling, the third index value of each physiological index is obtained based on the adjusted sampling parameter; the prediction function parameter is updated based on the third index value; and the prediction function parameter includes the calculation parameter corresponding to blood oxygen, heart rate and respiration.

[0364] The calculation parameter is a parameter in the prediction function (such as the above formulas one to four) used in the above prediction process.

[0365] Optionally, the PPG signal is obtained by controlling the sensor to sample according to the adjusted sampling parameters, and then the third index value of each physiological index is obtained according to the PPG signal, and the prediction function parameters are updated accordingly. That is, the prediction function parameters are updated in real time based on the corresponding index values after sampling through the multi-feature network control, so as to further improve the accuracy of the subsequent obtained index values.

[0366] The above method monitors the physiological index through the PPG sensor, reduces the power consumption of the sensor, and determines the target sampling parameters corresponding to the minimum value obtained under the condition of joint power consumption calculation and error bound calculation based on each prediction value, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient, and a current characteristic coefficient of the light source of the PPG sensor, and adjusts the sampling parameters of the sampling controller based on the target sampling parameters. Thus, the power consumption of the sensor is reduced, and the error in monitoring the physiological index is reduced, thereby reducing the power consumption of the sensor while ensuring the estimation accuracy.

[0367] Further, the target sampling parameters, the plurality of sampling frequencies, and the plurality of emission intensities are determined in the following manner. When the target sampling parameters corresponding to the minimum value obtained under the condition of joint calculation are determined, the error lower bound is taken as the optimization target, the end time of the first time interval, the sampling value at the end time, and the first blood oxygen content are used to determine each sampling time, and then the sampling frequency corresponding to each sampling time and the emission intensity of the emission unit of the PPG sensor are determined.

[0368] The error lower bound refers to the minimum difference between the calculation result and the true value, in other words, the minimum prediction error under given conditions. In this embodiment, the calculation result refers to each sampling time.

[0369] The prediction values at least include the second time interval of the respiratory movement in the next respiratory cycle at a relatively static state, and the sampling times at least include the first time corresponding to the maximum signal intensity of the next respiratory cycle and at least one second time within the second time interval. That is, the sampling times at least include the time within the second time interval corresponding to the heart rate or blood oxygen of the next respiratory cycle, and the time within the time interval corresponding to the respiration of the next respiratory cycle, so as to consider the signal characteristics of the PPG signal in different characteristic time intervals, and improve the accuracy of each sampling time.

[0370] It should be noted that the target sampling parameters are determined at the end of the current respiratory cycle, that is, at the end time of the first time interval.

[0371] For example, the sampling time is determined, that is, the timing signal of the next respiratory cycle is predicted, and the sampling time can be determined by the following Formula Six: For example, the sampling time is determined, that is, the timing signal of the next respiratory cycle is predicted, and the sampling time can be determined by the following Formula Six:

[0372] wherein i is the serial number of the respiratory cycle, p i is the predicted probability, l(·) is a probability calculation formula, s i is the sampling time, μ i is the mean value of the PPG signal intensity in the prediction interval, σ i is the signal distribution of the time series signal.

[0373] Correspondingly, at the first sampling time t i1 after the end time of the first time interval, the predicted value can be calculated by the formula It should be noted that the distribution parameters in the formula, including μ i1 and σ i1 are affected by the blood oxygen content, so before the prediction sampling time t i1 , the distribution parameters need to be updated by the following formula six and formula seven: μ i1 = μ i0 + z1(Spo2) (seven) σ i1 = σ i0 + z2(Spo2) (eight)

[0374] wherein μ i0 is the mean value of the end time of the first time interval, that is, the mean value of the last sampling time of the sampling time t i1 ; σ i0 is the signal distribution of the time series signal at the end time of the first time interval; z1(Spo2) and z2(Spo2) are the normal distribution of the blood oxygen content.

[0375] As shown in FIG. 3, t i0 is the end time of the first time interval, and each sampling time can include t i1 -t i5 , and the solid line in the prediction interval is drawn by the mean value corresponding to each sampling time.

[0376] In an exemplary embodiment, the above method further includes steps 1202-1208. Wherein:

[0377] Step 1202, obtaining a second PPG signal sampled by the PPG sensor in the next respiratory cycle, and a second time interval in which the predicted respiratory movement is relatively static.

[0378] Step 1204, determining a third time interval in the next respiratory cycle except the second time interval.

[0379] Step 1206, filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to each physiological indicator.

[0380] In step 1208, the second index value of each physiological index is estimated based on the filtered second PPG signal.

[0381] Each physiological index has a different characteristic signal, i.e., a different feature on the PPG signal, which corresponds to the prior parameter of the characteristic signal.

[0382] For example, in order to avoid loss of information, the signal features corresponding to the second time interval and the third time interval are filtered by the prior parameters corresponding to the physiological index, rather than being filtered multiple times by the prior parameters corresponding to all physiological indexes.

[0383] For example, the at least two physiological indexes are respiration and blood oxygen, respectively. When filtering the second PPG signal, the second PPG signal is filtered in the second time interval corresponding to respiration by using the prior parameter corresponding to respiration, and the second PPG signal is filtered in the third time interval corresponding to blood oxygen by using the prior parameter corresponding to blood oxygen. For example, the second PPG signal has a time interval 1 and a time interval 2, wherein the time interval 1 corresponds to respiration and the time interval 2 corresponds to blood oxygen. The prior parameter corresponding to respiration is used to filter the signal feature of the second PPG signal corresponding to the time interval 1, and the prior parameter corresponding to blood oxygen is used to filter the signal feature of the second PPG signal corresponding to the time interval 2.

[0384] It should be noted that after filtering the second PPG signal, the characteristic PPG signal corresponding to each physiological index can be extracted therefrom.

[0385] It should be noted that when calculating the index values corresponding to heart rate and blood oxygen, the signal feature corresponding to the time interval of the respiratory motion in the relatively static state in the second PPG signal is generally used for calculation. The accuracy of the index value calculated by using the signal feature is relatively high.

[0386] Therefore, the third time interval corresponding to heart rate and blood oxygen can be the same, and the third time interval is a time interval in a respiratory cycle except for the second time interval corresponding to respiration. As shown in FIG. 4, FIG. 4 is a waveform of the characteristic PPG signal corresponding to respiration, wherein the first time interval corresponding to respiration includes t o_start -t o_stop and t i_start -t i_stop The first time interval corresponding to heart rate and blood oxygen is t o_stop -t i_start .

[0387] In an embodiment, if there is a same part in the time interval corresponding to each of the at least two physiological indicators, the same part of the time interval can be divided into intervals in a certain proportion, so that the corresponding physiological indicators correspond to different divided intervals respectively. For example, there is a same part of time interval 3 in the time interval 1 corresponding to respiration and the time interval 2 corresponding to blood oxygen. The time interval 3 is divided into two parts in an equal manner, and each part corresponds to respiration and blood oxygen respectively.

[0388] In another embodiment, for each physiological indicator, the second PPG signal is filtered in the second time interval and the third time interval based on the prior parameter corresponding to the physiological indicator, which specifically includes:

[0389] If there are two target physiological indicators corresponding to the second time interval in the at least two physiological indicators, the correlation between each sampling data corresponding to the second time interval in the second PPG signal and each target physiological indicator is determined respectively; each sampling data is filtered based on the prior parameter corresponding to the target physiological indicator with the highest correlation.

[0390] It should be noted that the second PPG signal is actually composed of a plurality of sampling data, and the waveform is connected by points corresponding to the sampling values of the sampling data.

[0391] The correlation refers to the matching degree between the sampling data and the prior parameter corresponding to each physiological indicator.

[0392] Exemplarily, each sampling data and different physiological indicators have different correlations. When each sampling data is filtered based on the prior parameter corresponding to the physiological indicator with the highest correlation, more information of the sampling data can be retained, thereby improving the accuracy of the estimated first indicator value.

[0393] In this embodiment, when the indicator value is determined by the PPG signal, the characteristics corresponding to different physiological indicators are considered to be related to each other, and the overall PPG signal is not filtered respectively, but is segmented and filtered by the corresponding time interval. That is, the PPG signal corresponding to each time interval is only filtered by the physiological indicator corresponding to the physiological indicator, but not filtered by all physiological indicators, thereby avoiding the loss of information in the PPG signal and improving the accuracy of each indicator value.

[0394] On the basis of each of the above embodiments, the second indicator values include a second heart rate value, a second blood oxygen content, and a second respiration characteristic value, and the second indicator values of each physiological indicator estimated based on the filtered second PPG signal include the following steps 1302 to 1308. Wherein:

[0395] Step 1302, respectively extracting a respiration feature PPG signal corresponding to respiration, an oxygen feature PPG signal corresponding to oxygen and a heart rate feature PPG signal corresponding to heart rate from the filtered second PPG signal.

[0396] Wherein, each feature PPG signal is estimated by correlation degree calculation of the filtered second PPG signal and the corresponding feature signal, that is, the respiration feature signal can be estimated by the respiration feature PPG signal, the oxygen feature signal can be estimated by the oxygen feature PPG signal, and the heart rate feature signal can be estimated by the heart rate feature PPG signal.

[0397] Step 1304, determining a second respiration feature value based on the respiration feature PPG signal; the second respiration feature value includes a second respiration signal feature value and a fourth time interval of respiration movement in relative static state.

[0398] Wherein, the second respiration signal feature value includes a second period, a second amplitude and a second phase.

[0399] Optionally, the second respiration feature value can be directly identified from the waveform corresponding to the respiration feature PPG signal.

[0400] As shown in FIG. 4, FIG. 4 is a waveform diagram corresponding to the respiration feature PPG signal. For Ti period, the exhalation interval t o_start -t o_stop , the inhalation interval t i_start -t i_stop , and t o_stop -t i_start are identified from the respiration feature PPG signal.

[0401] Step 1306, determining a second oxygen content based on the fourth time interval and the oxygen feature PPG signal.

[0402] As shown in FIG. 5, the fourth time interval is fused with the oxygen feature PPG signal, that is, the signal feature of the oxygen feature PPG signal corresponding to the fourth time interval in the oxygen feature PPG signal, that is, t o_stop -t i_start , is determined. Then the second oxygen content is determined through the signal feature.

[0403] Further, when the second oxygen content is determined through the signal feature, the sampling data is obtained from the signal feature; then the second oxygen content is calculated through the sampling data.

[0404] The calculation process is: determining the first alternating component and the first direct current component of the red light signal and the second alternating component and the second direct current component of the infrared signal in the sampling data; obtaining the second blood oxygen content based on the first alternating component, the first direct current component, the second alternating component and the second direct current component.

[0405] Referring to formula nine, the second blood oxygen content Spo2 can be calculated by the formula nine:

[0406] Wherein, Ac red is the first alternating component of the red light signal, Dc red is the first direct current component of the red light signal, Ac ir is the second alternating component of the infrared signal, Dc ir is the second direct current component of the infrared signal.

[0407] Step 1308, determining the second heart rate value based on the fourth time interval and the heart rate characteristic PPG signal.

[0408] Similarly, as shown in FIG. 5, the fourth time interval is fused with the heart rate characteristic PPG signal, that is, the signal characteristic of the heart rate characteristic PPG signal corresponding to the fourth time interval in the heart rate characteristic PPG signal, that is, t o_stop -t i_start corresponding signal characteristic. Then the second heart rate value is determined through the signal characteristic.

[0409] In an exemplary embodiment, after determining the target sampling parameter of the next breathing cycle corresponding to the PPG sensor in the lowest power consumption and the highest accuracy based on the predicted value, the preset accuracy optimization coefficient, the preset power consumption optimization coefficient and the current characteristic coefficient of the light source of the PPG sensor, the method further comprises:

[0410] Updating the prior parameter corresponding to each physiological index based on the target sampling parameter and the predicted value.

[0411] Wherein, the prior parameter includes but is not limited to the index value and the sampling parameter corresponding to the physiological index. For example, the period corresponding to the respiration and the time interval of the respiration movement in the relative static state.

[0412] It should be noted that the second PPG signal of the next breathing cycle can be filtered based on the updated prior parameter, so as to improve the estimation accuracy of the third index value corresponding to at least two physiological indexes in the next breathing cycle respectively.

[0413] In an exemplary embodiment, as shown in FIG. 12, the overall flow of the method can include steps 1-19. Wherein:

[0414] Step 1: Control the PPG sensor (including a PPG emitting unit and a PPG receiving unit) to sample at a sampling time, a sampling frequency and an emission intensity to obtain a PPG signal, and input the PPG signal into a PPG feature extractor. The PPG sensor as a respiratory monitoring node can be attached to the chest of a monitored object or placed in the chest cavity of the monitored object.

[0415] Step 2: Perform priori processing and filtering on the PPG signal.

[0416] Step 3: Extract a respiratory feature PPG.

[0417] Step 4: Extract a blood oxygen feature PPG.

[0418] Step 5: Extract a heart rate feature PPG.

[0419] Step 6: Perform respiratory motion detection based on the respiratory feature PPG to obtain a first respiratory signal feature value and a first time interval of the respiratory motion in a relatively static state.

[0420] Step 7: Perform blood oxygen content calculation based on the blood oxygen feature PPG and the first time interval obtained in subsequent step 9.

[0421] Step 8: Perform heart rate value calculation based on the heart rate feature PPG and the first time interval obtained in subsequent step 9.

[0422] Step 9: Input the first time interval into a blood oxygen content calculation module and a heart rate calculation module, so that the blood oxygen content calculation module calculates a first blood oxygen content and the heart rate calculation module calculates a first heart rate value.

[0423] Step 10: Input the first time interval into a respiratory prediction module; the respiratory prediction module performs respiratory prediction based on the first time interval and a first heart rate value obtained in subsequent step 12 to obtain a second time interval of respiratory motion in a relatively static state in a next respiratory cycle; and the respiratory prediction module predicts a second respiratory signal feature value in the next respiratory cycle based on a first blood oxygen content obtained in step 11, the first heart rate value obtained in subsequent step 12 and the first time interval.

[0424] Step 11: Input the first blood oxygen content into a blood oxygen prediction module, so that the blood oxygen prediction module predicts a second blood oxygen content in the next respiratory cycle based on the first blood oxygen content and a second respiratory signal feature value obtained in subsequent step 13.

[0425] Step 12: Input the first heart rate value into the respiratory prediction module.

[0426] Step 13: The respiratory prediction module inputs the second respiratory signal feature value into the blood oxygen prediction module.

[0427] Step 14: input the second respiratory signal feature value and the second time interval predicted by the respiratory prediction module into the energy factor module; the energy factor module inputs the preset power consumption optimization coefficient, the current characteristic coefficient, the second respiratory signal feature value and the second time interval into the respiratory monitoring optimizer through step 16. The energy factor module also obtains the predicted value of the other physiological indicators of the next respiratory cycle, which is also input into the respiratory monitoring optimizer through step 16. The respiratory monitoring optimizer performs joint optimization based on the information received through steps 16 and 17 to obtain the target sampling parameter.

[0428] Step 15: input the second respiratory signal feature value and the second time interval predicted by the respiratory prediction module into the accuracy factor module; the accuracy factor module inputs the preset accuracy optimization coefficient, the second respiratory signal feature value and the second time interval into the respiratory monitoring optimizer through step 17. The accuracy factor module also obtains the predicted value of the other physiological indicators of the next respiratory cycle, which is input into the respiratory monitoring optimizer through step 17. The respiratory monitoring optimizer performs joint optimization based on the information received through steps 16 and 17 to obtain the target sampling parameter.

[0429] Step 18: input the predicted value obtained in the prediction process, such as the second time interval, into the PPG feature extractor.

[0430] Step 19: input the target sampling parameter into the sampling controller; the sampling controller adjusts the sampling parameter, and obtains the third indicator value of each physiological indicator based on the adjusted sampling parameter in the next respiratory cycle; update the prediction function parameter based on the third indicator value; the prediction function parameter includes the calculation parameter corresponding to blood oxygen, heart rate and respiration.

[0431] It should be understood that although each step in the flowchart involved in each embodiment described above is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0432] Based on the same inventive concept, the embodiments of the present application also provide a computer device for implementing the above method. The implementation scheme for solving the problem provided by the computer device is similar to the implementation scheme described in the above method, so the specific limitations in one or more computer device embodiments provided below can refer to the limitations of the above method described above, which will not be repeated here.

[0433] In an exemplary embodiment, a computer device is provided, which can be a respiratory monitoring device, and an internal structure diagram of the computer device can be as shown in FIG. 13. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through Bluetooth, WIFI, mobile cellular network, Near Field Communication (NFC) or other technologies. The computer program is executed by the processor to implement the above method.

[0434] Those skilled in the art can understand that the structure shown in FIG. 13 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0435] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0436] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magneto resistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0437] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, it should be considered as the scope of the present application.

[0438] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method of optimizing accuracy of respiratory monitoring, the method comprising: The method comprises: acquiring first index values corresponding to at least two physiological indexes respectively; each first index value is estimated based on a first PPG signal obtained by sampling a PPG sensor under control of a sampling controller; predicting a predicted value of each physiological index in a next breathing cycle based on each first index value; determining a target sampling parameter of the next breathing cycle at a highest estimation accuracy based on each first index value and each predicted value; the estimation accuracy is an accuracy when each second index value of the next breathing cycle is estimated based on a second PPG signal corresponding to the next breathing cycle; adjusting the sampling parameter of the sampling controller based on the target sampling parameter, and obtaining third index values of each physiological index based on the adjusted sampling parameter when sampling in the next breathing cycle, and updating a prediction function parameter based on the third index values; the prediction function parameter comprises calculation parameters corresponding to blood oxygen, heart rate and respiration.

2. The method of claim 1, wherein, Each first index value comprises a first heart rate value, a first blood oxygen content and a first respiration characteristic value, the first respiration characteristic value comprises a first time interval when respiration movement is relatively static, and the target sampling parameter further comprises a plurality of sampling time points; the determination of the target sampling parameter of the next breathing cycle at the highest estimation accuracy based on each first index value and each predicted value comprises: determining each sampling time point based on an end time point of the first time interval, a sampling value at the end time point and the first blood oxygen content, with an error estimation lower bound as an optimization target; determining a sampling frequency of each sampling time point and an emission intensity of an emission unit of the PPG sensor in the next breathing cycle at the highest estimation accuracy based on each first index value and each predicted value respectively.

3. The method of claim 2, wherein, Each predicted value at least comprises a second time interval when respiration movement is relatively static in the next breathing cycle, and each sampling time point at least comprises a first time point corresponding to a maximum signal intensity of the next breathing cycle and at least one second time point in the second time interval.

4. The method of claim 1, wherein, Each first index value comprises a first heart rate value, a first blood oxygen content and a first respiration characteristic value, the first respiration characteristic value comprises a first respiration signal characteristic value and a first time interval when respiration movement is relatively static, and the prediction of the predicted value of each physiological index in the next breathing cycle based on each first index value comprises: predicting a second respiration signal characteristic value in the next breathing cycle based on the first heart rate value, the first blood oxygen content and the first respiration signal characteristic value; predicting a second blood oxygen content in the next breathing cycle based on the first blood oxygen content and the second respiration signal characteristic value; predicting a second time interval when respiration movement is relatively static in the next breathing cycle based on the first time interval and the first heart rate value.

5. The method of claim 1, wherein, The method further comprises: acquiring a second PPG signal obtained by sampling a PPG sensor in a next breathing cycle, and a second time interval when respiration movement is relatively static which is predicted; determining a third time interval in the next breathing cycle except the second time interval; filtering the second PPG signal in the second time interval and the third time interval based on the prior parameters corresponding to the physiological indicators; estimating second indicator values of the physiological indicators based on the filtered second PPG signal.

6. The method of claim 5, wherein, The filtering the second PPG signal in the second time interval and the third time interval based on the prior parameters corresponding to the physiological indicators comprises: if there are two target physiological indicators corresponding to the second time interval in the at least two physiological indicators, determining the correlation between each sampling data corresponding to the second time interval in the second PPG signal and each target physiological indicator; filtering each sampling data based on the prior parameter corresponding to the target physiological indicator with the highest correlation.

7. The method of claim 5 or claim 6, wherein, The second indicator values comprise a second heart rate value, a second blood oxygen content, and a second respiratory characteristic value, and the estimating second indicator values of the physiological indicators based on the filtered second PPG signal comprises: extracting a respiratory characteristic PPG signal corresponding to respiration, a blood oxygen characteristic PPG signal corresponding to blood oxygen, and a heart rate characteristic PPG signal corresponding to heart rate from the filtered second PPG signal; determining the second respiratory characteristic value based on the respiratory characteristic PPG signal, wherein the second respiratory characteristic value comprises a second respiratory signal characteristic value and a fourth time interval of respiratory movement in relative static state; determining the second blood oxygen content based on the fourth time interval and the blood oxygen characteristic PPG signal; determining the second heart rate value based on the fourth time interval and the heart rate characteristic PPG signal.

8. A method of respiratory monitoring power consumption optimization, the method comprising: The method comprises: obtaining first indicator values corresponding to a current respiratory cycle and second indicator values corresponding to a previous respiratory cycle of at least two physiological indicators, and a power consumption characteristic coefficient of a light source of a PPG sensor, wherein each first indicator value is estimated based on a first PPG signal obtained by sampling the PPG sensor under the control of a sampling controller, and the light source comprises a red light source and an infrared light source; predicting a predicted value of each physiological indicator in a next respiratory cycle based on each first indicator value; determining a target sampling parameter of the next respiratory cycle corresponding to the lowest power consumption of the PPG sensor based on the power consumption characteristic coefficient, the second indicator values, and the predicted value; adjusting the sampling parameter of the sampling controller based on the target sampling parameter, obtaining third indicator values of the physiological indicators in the next respiratory cycle based on the adjusted sampling parameter, and updating a prediction function parameter based on the third indicator values, wherein the prediction function parameter comprises calculation parameters corresponding to blood oxygen, heart rate, and respiration.

9. The method of claim 8, wherein, The first indicator values comprise a first heart rate value, a first blood oxygen content, and a first respiratory characteristic value, the first respiratory characteristic value comprises a first respiratory signal characteristic value and a first time interval of respiratory movement in relative static state, and the predicting a predicted value of each physiological indicator in a next respiratory cycle based on each first indicator value comprises: predict a third respiration signal feature value of a next respiration cycle based on the first heart rate value, the first blood oxygen content, and the first respiration signal feature value; predict a third blood oxygen content of a next respiration cycle based on the first blood oxygen content and the third respiration signal feature value; predict a third time interval of respiration movement in relative static state of a next respiration cycle based on the first time interval and the first heart rate value.

10. The method of claim 9, wherein, The prediction of the third time interval of respiration movement in relative static state of a next respiration cycle based on the first time interval and the first heart rate value comprises: determine a first gain of a first start time and a second gain of a first end time of the first time interval based on the first heart rate value respectively; determine a third start time of the third time interval of respiration movement in relative static state of a next respiration cycle based on the first start time and the first gain; determine a third end time of the third time interval of respiration movement in relative static state of a next respiration cycle based on the first end time and the second gain.

11. The method of claim 8, wherein, The acquisition of the first index value of each of the at least two physiological indexes corresponding to the current respiration cycle comprises: acquire a first PPG signal sampled by the PPG sensor in the current respiration cycle, and a fourth time interval of respiration movement in relative static state of the current respiration cycle predicted in a previous respiration cycle; determine a fifth time interval of the current respiration cycle except the fourth time interval; filter the first PPG signal in the fourth time interval and the fifth time interval based on the prior parameters corresponding to each of the physiological indexes respectively; estimate the first index value corresponding to each of the physiological indexes based on the filtered first PPG signal.

12. The method of claim 11, wherein, The at least two physiological indexes comprise respiration, blood oxygen, and heart rate, and the first index value of each of the physiological indexes comprises a first respiration feature value, a first blood oxygen content, and a first heart rate value, and the estimation of the first index value corresponding to each of the physiological indexes based on the filtered first PPG signal comprises: extract a respiration feature PPG signal corresponding to respiration, a blood oxygen feature PPG signal corresponding to blood oxygen, and a heart rate feature PPG signal corresponding to heart rate from the filtered first PPG signal respectively; determine the first respiration feature value based on the respiration feature PPG signal; the first respiration feature value comprises a first respiration signal feature value and a first time interval of respiration movement in relative static state; determine the first blood oxygen content based on the first time interval and the blood oxygen feature PPG signal; determine the first heart rate value based on the first time interval and the heart rate feature PPG signal.

13. The method of claim 12, wherein, The determination of the first blood oxygen content based on the first time interval and the blood oxygen feature PPG signal comprises: determine a signal feature corresponding to the first time interval in the blood oxygen feature PPG signal; determine the first blood oxygen content based on the signal feature.

14. The method of claim 11, wherein, The method further comprises, after the determination of the target sampling parameter of the PPG sensor corresponding to a next respiration cycle in the lowest power consumption based on the power consumption characteristic coefficient, the second index value, and the predicted value: Update a prior parameter corresponding to each of the physiological indicators based on the target sampling parameter and the prediction value.

15. A respiratory monitoring joint optimization method, comprising: The method comprises: Obtaining first indicator values corresponding to at least two physiological indicators in a current breathing cycle respectively, each of the first indicator values being estimated based on a first PPG signal obtained by sampling a PPG sensor under control of a sampling controller; Predicting a prediction value of each of the physiological indicators in a next breathing cycle based on each of the first indicator values; Determining a target sampling parameter corresponding to a minimum value obtained under joint calculation of power consumption and error bound based on each of the prediction values, a preset accuracy optimization coefficient, a preset power consumption optimization coefficient, and a current characteristic coefficient of a light source of the PPG sensor; the light source comprises a red light source and an infrared light source; the target sampling parameter is used for controlling the PPG sensor to sample in the next breathing cycle under control of the sampling controller; Adjusting a sampling parameter of the sampling controller based on the target sampling parameter.

16. The method of claim 15, wherein, The determining of the target sampling parameter corresponding to the minimum value obtained under the joint calculation of power consumption and error bound based on each of the prediction values, the preset accuracy optimization coefficient, the preset power consumption optimization coefficient, and the current characteristic coefficient of the light source of the PPG sensor comprises: Performing power consumption calculation based on each of the prediction values, and performing error bound calculation based on each of the prediction values; Performing joint calculation by taking the preset power consumption optimization coefficient and the current characteristic coefficient of the light source of the PPG sensor as first coefficients in the power consumption calculation, and taking the preset accuracy optimization coefficient as a second coefficient in the error bound calculation; Determining the target sampling parameter corresponding to the minimum value obtained under the joint calculation.

17. The method of claim 16, wherein, Each of the first indicator values comprises a first heart rate value, a first blood oxygen content, and a first breathing characteristic value, the first breathing characteristic value comprises a first time interval when breathing movement is relatively static, the target sampling parameter comprises a plurality of sampling time points, a plurality of sampling frequencies, and a plurality of emission intensities, and the determining of the target sampling parameter corresponding to the minimum value obtained under the joint calculation comprises: Determining each of the sampling time points based on an end time point of the first time interval, a sampling value at the end time point, and the first blood oxygen content, with a lower bound of error estimation as an optimization target; Determining a sampling frequency corresponding to each of the sampling time points and an emission intensity of an emission unit of the PPG sensor.

18. The method of claim 17, wherein, Each of the prediction values comprises at least a second time interval when breathing movement is relatively static in the next breathing cycle, and each of the sampling time points comprises at least a first time point corresponding to a maximum signal intensity in the next breathing cycle and at least one second time point in the second time interval.

19. The method of claim 15, wherein, Each of the first indicator values comprises a first heart rate value, a first blood oxygen content, and a first breathing characteristic value, the first breathing characteristic value comprises a first breathing signal characteristic value and a first time interval when breathing movement is relatively static, and the predicting of the prediction value of each of the physiological indicators in the next breathing cycle based on each of the first indicator values comprises: Predicting a second breathing signal characteristic value in the next breathing cycle based on the first heart rate value, the first blood oxygen content, and the first breathing signal characteristic value; predict a second blood oxygen content of a next breathing cycle based on the first blood oxygen content and the second respiratory signal feature value; predict a second time interval of the respiratory movement in relative static state of the next breathing cycle based on the first time interval and the first heart rate value.

20. The method of claim 15, wherein, The method further comprises: obtaining a second PPG signal sampled by the PPG sensor in the next breathing cycle and the second time interval of the respiratory movement in relative static state predicted; determining a third time interval of the next breathing cycle except the second time interval; filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators; estimating second indicator values of the physiological indicators based on the filtered second PPG signal.

21. The method of claim 20, wherein, The filtering the second PPG signal in the second time interval and the third time interval respectively based on the prior parameters corresponding to the physiological indicators comprises: if there are two target physiological indicators corresponding to the second time interval in the at least two physiological indicators, determining a correlation between each sampling data corresponding to the second time interval in the second PPG signal and each target physiological indicator respectively; filtering each sampling data based on the prior parameter corresponding to the target physiological indicator with the highest correlation.

22. The method of claim 15, wherein, The sampling parameter adjustment of the sampling controller based on the target sampling parameter comprises: obtaining third indicator values of the physiological indicators based on the adjusted sampling parameters when sampling in the next breathing cycle; updating the prediction function parameters based on the third indicator values; the prediction function parameters comprise calculation parameters corresponding to blood oxygen, heart rate and respiration. 23.A computer device, comprising a memory and a processor, wherein the memory stores a computer program. The processor executes the computer program to implement the steps of the method in any one of claims 1 to 22.

24. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 22.

Images