CN106175739B - Phase measurement and optimization method and system between a kind of pulse R R during sleep - Google Patents

Abstract

The present invention relates to phase measurement and optimization method and system between the pulse R R during a kind of sleep, the sleep quality of testee is detected in real time by the high equipment of untouchable and portable degree and accuracy in detection high method, only need by detecting pulse signal, the extraction of time value between accurate R R is realized with many algorithms, time value carrying out obtaining preliminary R R after screening for the first time pulse peak value by the covariance matching of pulse peak value, secondary investigation is carried out to time value unusual preliminary R R by AR models again, the effect of time value between the existing R R gathered out by PSG is reached, and it disclosure satisfy that user is in the demand of the optimization of phase between progress R R under the daily life styles such as house or travel, and reduce time and money consumption.

Description

A method and system for measuring and optimizing pulse R-R interval during sleep

technical field

The present invention relates to an R-R interval optimization technology applicable to heart rate variability in daily life environment, in particular to a non-contact way to monitor breathing, pulse and other vital signs during sleep to optimize the R-R interval Pulse R-R interval measurement and optimization method and system during sleep.

Background technique

According to the "China Cardiovascular Disease Report (2015)", it is estimated that there are currently 290 million cardiovascular disease patients in China, and 3.5 million people die of cardiovascular disease every year. With the improvement of living standards, the acceleration of the pace of life and the arrival of the global aging society, the morbidity and mortality of heart disease will increase. It can be seen that heart disease has become one of the main diseases threatening human beings. Modern medical research has shown that obtaining continuous and accurate R-R intervals for heart rate variability analysis can quickly understand the functional state of the heart.

In order to obtain accurate R-R interval values, the most recognized method is polysomnography (PSG) used in hospitals, but PSG is difficult to popularize and apply due to the complexity of the process, high cost, and poor user experience. Therefore, the study of a convenient and accurate detection method to replace PSG will provide strong technical support for the optimization of R-R interval and the analysis of heart rate variability, and has high practical value.

At present, there are roughly the following methods for obtaining the R-R interval:

1. Patent document CN104398257A discloses a periodic statistical analysis method of electrocardiogram waveforms, which reads a number of given cycles of electrocardiogram waveforms, distinguishes the position of each cycle of electrocardiogram QRS complexes and calculates the position of all cycle electrocardiogram waveforms. Overlay the mean plots to determine the R-R interval value. This type of method needs to use PSG to measure the ECG signal first. Therefore, the process involved is more complicated, the cost is expensive, and the user experience is poor. Secondly, this type of method can only do offline analysis of a piece of data, and cannot achieve continuous Real-time detection.

2. A document from Ocean University of China discloses a method for determining the R-R interval by detecting R waves, obtaining data from the MIT-BIH ECG signal database and performing preprocessing such as filtering on the original signal, and then using threshold detection The algorithm detects R waves and predicts subsequent R-R interval values. This type of method can only be analyzed offline, and cannot obtain real-time ECG signals, and the calculation of the R-R interval value can only stay in theory, and has poor practicability.

3. Patent document CN103479349A provides a method and system for acquiring and processing ECG signal data, using the R-R interval processing unit to make the signal bandwidth of the R-R interval fall within the frequency bandwidth required by the Hilbert transform for the best amplitude-frequency characteristics Within, and through resampling, filtering, etc. to solve missed detection, false detection and other problems. Such methods interfere too much with the calculation of the original ECG signal and the R-R interval, and the artificial correction is excessive, and the accuracy of the calculated R-R interval value needs to be considered.

In short, the existing optimization methods for R-R interval value either need to rely on PSG, which is complex and cumbersome and has extremely low user experience, and can only be detected and checked in a hospital environment, which cannot be applied to daily life environments such as family or business trips; or cannot be done. to real-time collection and detection, and the accuracy of the calculated R-R interval value is not high.

Contents of the invention

The purpose of the present invention is to solve the shortcomings and deficiencies of the above-mentioned prior art, and to provide a method and system for measuring and optimizing the pulse R-R interval during sleep, through non-contact and highly portable equipment and a method with high detection accuracy Real-time detection of the subject's sleep condition, and only need to detect the pulse signal, use a variety of algorithms to achieve accurate R-R interval value extraction, and obtain the preliminary R-R interval value after initial screening of the pulse peak value through the covariance matching of the pulse peak value Interval value, and then use the AR model to conduct a secondary investigation on the singular preliminary R-R interval value, which achieves the effect of the existing R-R interval value collected by PSG, and can meet the needs of users in daily life such as home or travel. Under the demand of R-R interval optimization, and reduce time and money consumption.

In order to achieve the above object, the technical scheme adopted in the present invention is as follows:

At first, the present invention provides a kind of pulse R-R interval measurement and optimization method during sleep, and it comprises the following steps:

Step 1: Obtain and calculate the accurate pulse rate, locate the pulse peak point, and accurately obtain the pulse waveform;

Step 2: Calculate the preliminary R-R interval value according to the pulse waveform;

Step 3: Set a time interval, and establish an AR model used to represent the effective peak information in the time interval according to the preliminary R-R interval value in the time interval;

Step 4: The AR model calculates the R-R interval value of the N+1th point according to the preliminary R-R interval value of the first N points in the time interval;

Step 5: Calculate the difference between the R-R interval value of the N+1th point calculated by the AR model and the actual R-R interval value of the N+1th point in the time interval; judge Whether the difference falls within a threshold range, if yes, record the actual R-R interval value of the N+1 point in the time interval; otherwise, consider the detection failure and discard the Nth point in the time interval +1 point of actual R-R interval value;

Step 6: Sequentially extract a plurality of time intervals with the same duration as the time interval at equal time intervals, and repeat steps 3 to 5; until all time intervals are processed, the R-R interval during sleep is optimized.

As a further improvement of the present invention, the step 1 specifically includes the following steps:

Step 11: collect sleep signals in real time;

Step 12: Process the sleep signal to obtain the pulse rate;

Step 13: Perform morphological filtering, peak detection, and covariance matching processing on the pulse rate in sequence to obtain accurate pulse rate data, and generate a pulse waveform according to the pulse rate data.

As a further improvement of the present invention, the step 11 is to realize the real-time collection of sleep signals through a piezoelectric sensor, which is placed on the bottom of the subject's pillow during collection.

As a further improvement of the present invention, the type of the sleep signal collected by the piezoelectric sensor in the step 11 is an analog signal type, and the analog signal is sent to an A/D conversion module, which is converted into a digital signal by the A/D conversion module Types of sleep signals.

As a further improvement of the present invention, the step 2 specifically includes the following steps:

Step 21: Extract a segment of the pulse signal from the pulse waveform, and create a pulse waveform pulse in the segment of the pulse signal;

Step 22: moving and matching the pulse waveform pulse with the pulse signal to obtain a set of covariance matching correlation functions;

Step 23: Establishing a moving window with a period of 6s in the correlation function, and moving the moving window with a period of 1s in the correlation function, and selecting the maximum value point in the moving window after each movement, And all the selected maximum value points are used as the pulse fluctuation pulse in the pulse signal;

Step 24: According to the characteristic that the heart rate does not have a sudden change in a short period of time, judge the pulse period of the pulse fluctuation pulse formed in step 23, remove the abnormal peak value, and obtain effective peak value data;

Step 25: mark the obtained effective peak data as the pulse peak point in the correlation function, and calculate the preliminary R-R interval value.

As a further improvement of the present invention, in the step 3, a time interval of 6s is set, and according to 4 to 6 preliminary R-R interval values Xi within the time interval, an effective AR model of peak information, where the value of i is an integer between 1 and 6, including 1 and 6.

In the step 4, the AR model estimates the RR interval value X of the fourth point according to the preliminary RR interval values X ₁ , X ₂ and X ₃ of the first to third points in the time interval ₄ , X ₄ =a×X ₃ +a ² ×X ₂ +a ³ ×X ₁ ; where the value of a is the real number solution of the equation a+a ² +a ³ =1.

In the step 5, the following steps are specifically included:

Step 51: Calculate the difference between the RR interval value _X4 of the fourth point calculated by the AR model and the actual RR interval value Y4 of the _fourth point within the time interval;

Step 52: Determine whether the absolute value |Y ₄ -X ₄ | of the difference is within X ₄ ×20%, if yes, record the actual RR interval value Y ₄ of the fourth point in the time interval ; Otherwise, the detection is regarded as failure, and the actual RR interval value Y ₄ of the fourth point in the time interval is discarded.

As a further improvement of the present invention, in the step 6, after the 6s time interval in the step 3, take 1s as the time interval, sequentially extract a plurality of 6s time intervals along the forward direction of the time axis, and repeat the Steps 3-5 above; until all time intervals are processed, the optimization of the R-R interval during sleep is realized.

In order to achieve another purpose of the present invention, the present invention also provides a pulse R-R interval measurement and optimization system during sleep corresponding to the above-mentioned pulse R-R interval measurement and optimization method during sleep. Electrical sensors, A/D conversion modules and processors;

The piezoelectric sensor is used to collect the sleep analog signal during the sleep of the subject in real time, and transmit it to the A/D conversion module;

The A/D conversion module is used to convert the sleep analog signal into a sleep digital signal and send it to the processor;

The processor is used to process the sleep digital signal to obtain a pulse rate waveform, and to process the pulse rate waveform to obtain a preliminary R-R interval value, and to check the initial R-R interval value to obtain an optimized R-R interval value to optimize the R-R interval during sleep.

Through the above technical scheme, the present invention has achieved the following beneficial technical effects:

The present invention detects the sleep condition of the subject in real time through a non-contact and highly portable device and a method with high detection accuracy, and only needs to detect the pulse signal and use various algorithms to realize the accurate R-R interval value. Extraction, the primary R-R interval value is obtained after the initial screening of the pulse peak value through the covariance matching of the pulse peak value, and then the singular preliminary R-R interval value is checked twice through the AR model, reaching the existing R-R interval value collected by PSG The effect of the interval value has high accuracy, and can meet the needs of users to optimize the R-R interval in daily life such as home or business trips, and reduce time and money consumption.

For better understanding and implementation, the present invention will be described in detail below in conjunction with the accompanying drawings.

Description of drawings

Fig. 1 is a brief flow chart of pulse R-R interval measurement and optimization method during sleep of the present invention;

Fig. 2 is the structural diagram of pulse R-R interval measurement and optimization system during sleep of the present invention;

FIG. 3 is a further improved structure diagram of the pulse R-R interval measurement and optimization system during sleep in FIG. 2 .

Detailed ways

Please refer to Fig. 1, the present invention provides a kind of pulse R-R interval measurement and optimization method during sleep,

At first, the present invention provides a kind of pulse R-R interval measurement and optimization method during sleep, and it comprises the following steps:

Step 1: Obtain and calculate an accurate pulse rate, locate the pulse peak point, and accurately obtain the pulse waveform; specifically, the step 1 includes the following steps:

Step 11: Real-time collection of sleep signals; in this step 11, the real-time collection of sleep signals is realized by a piezoelectric sensor, which is placed on the bottom of the subject's pillow during collection; and the sleep signal collected by the piezoelectric sensor The type is an analog signal type, which sends the analog signal to an A/D conversion module, and is converted into a digital signal type sleep signal by the A/D conversion module.

Step 12: Process the sleep signal to obtain the pulse rate; specifically, step 12 is to process the digital sleep signal to obtain the pulse rate;

Step 13: Perform morphological filtering, peak detection, and covariance matching processing on the pulse rate in sequence to obtain accurate pulse rate data, and generate a pulse waveform according to the pulse rate data.

Step 2: Calculate the preliminary R-R interval value according to the pulse waveform; specifically, the step 2 includes the following steps:

Step 21: Extract a segment of the pulse signal from the pulse waveform, and create a pulse waveform pulse in the segment of the pulse signal;

Step 22: moving and matching the pulse waveform pulse with the pulse signal to obtain a set of covariance matching correlation functions;

Step 23: Establishing a moving window with a period of 6s in the correlation function, and moving the moving window with a period of 1s in the correlation function, and selecting the maximum value point in the moving window after each movement, And all the selected maximum value points are used as the pulse fluctuation pulse in the pulse signal;

Step 24: According to the characteristic that the heart rate does not have a sudden change in a short period of time, judge the pulse period of the pulse fluctuation pulse formed in step 23, remove the abnormal peak value, and obtain effective peak value data; in this embodiment, the abnormal peak value The selection of the peak point is based on the two points before and after it. If there is a large gap between it and the two points before and after, and the mutation is large, the peak point is considered abnormal and excluded.

Step 25: mark the obtained effective peak data as the pulse peak point in the correlation function, and calculate the preliminary R-R interval value.

Step 3: set a time interval, according to the preliminary R-R interval value in the time interval, establish an AR model for representing the effective peak information in the time interval; in this embodiment, the step 3 In , a time interval of 6s is set, and an AR model used to represent the effective peak information in this time interval is established according to 4 to 6 preliminary R-R interval values Xi in this time interval, where the value of i is An integer between 1 and 6, including 1 and 6.

Step 4: The AR model calculates the RR interval value of the N+1th point according to the preliminary RR interval value of the first N points in the time interval; in this embodiment, in the step 4 , the AR model estimates the RR interval value X ₄ of the fourth point based on the preliminary RR interval values X ₁ , X ₂ and X ₃ of the first to third points in the time interval, X ₄ = a×X ₃ +a ² ×X ₂ +a ³ ×X ₁ ; where the value of a is the real number solution of the equation a+a ² +a ³ =1.

Step 5: Calculate the difference between the R-R interval value of the N+1th point calculated by the AR model and the actual R-R interval value of the N+1th point in the time interval; judge Whether the difference falls within a threshold range, if yes, record the actual R-R interval value of the N+1 point in the time interval; otherwise, consider the detection failure and discard the Nth point in the time interval The actual R-R interval value of +1 point; In the present embodiment, in described step 5, specifically comprise the following steps:

Step 51: Calculate the difference between the RR interval value _X4 of the fourth point calculated by the AR model and the actual RR interval value Y4 of the _fourth point within the time interval;

Step 52: Judging whether the absolute value |Y ₄ -X ₄ of the difference is within X ₄ ×20%, if yes, record the actual RR interval value Y ₄ of the fourth point in the time interval; Otherwise, it is regarded as a failed detection, and the actual RR interval value Y ₄ of the fourth point in the time interval is discarded.

Step 6: Sequentially extract a plurality of time intervals with the same duration as the time interval at equal time intervals, and repeat steps 3 to 5; until all time intervals are processed, the R-R interval during sleep is optimized. In this embodiment, in the step 6, after the 6s time interval in the step 3, a plurality of 6s time intervals are sequentially extracted along the forward direction of the time axis with 1s as the time interval, and the described Steps 3-5: until all the time intervals are processed, the optimization of the R-R interval during sleep is realized.

Please refer to Fig. 2, in addition, the present invention also provides a pulse R-R interval measurement and optimization system during sleep corresponding to the above-mentioned pulse R-R interval measurement and optimization method during sleep, the system includes piezoelectric sensors connected in sequence 10. An A/D conversion module 20 and a processor 30 .

The piezoelectric sensor 10 is used to collect sleep analog signals of the subject during sleep in real time and transmit them to the A/D conversion module 20 .

The A/D conversion module 20 is used for converting the sleep analog signal into a sleep digital signal and sending it to the processor 30 .

The processor 30 is used to process the sleep digital signal to obtain a pulse rate waveform, and process according to the pulse rate waveform to obtain a preliminary R-R interval value, and perform anomaly investigation on the preliminary R-R interval value to obtain an optimized R-R interval value. Interval value to optimize the R-R interval during sleep.

Please refer to Fig. 3, in order to prevent the piezoelectric sensor 10 from being affected by environmental factors and to prolong the service life of the piezoelectric sensor 10, as a better technical solution, the system for judging apnea during sleep of the present invention is also It includes a sensor mounting plate 40; the piezoelectric sensor 10 is embedded and installed in the middle of the sensor mounting plate 40, and its detection end is leakingly arranged on the surface of the sensor mounting plate 40, and is connected to the sensor mounting plate 40 The surface is even. In this embodiment, the length and width of the sensor mounting plate 40 are preferably consistent with or similar to the length and width of the subject's pillow.

In order to obtain a sleep digital signal with little external interference, as a better technical solution, the A/D conversion module 20 includes a filter circuit, an amplifier circuit and an A/D conversion circuit that are electrically connected in sequence and integrated on the same PCB board . The filtering circuit is electrically connected to the piezoelectric sensor 10, and the analog signal transmitted by the piezoelectric sensor 10 is filtered and then input to the amplifying circuit; the amplifying circuit amplifies the filtered analog signal, and Delivered to the A/D conversion circuit; the A/D conversion circuit is connected with the processor 30 through the serial port on the PCB board, which converts the analog signal through filtering and amplifying successively into a digital signal, and converts the digital signal through the serial port The signal is sent to the processor 30, and the processor 30 analyzes and processes the digital signal.

When the pulse signal needs to be detected, the entire sensor mounting plate 40 is placed on the bottom surface of the pillow that the subject needs to use when sleeping, or inserted into the pillow cover of the pillow, and pressed together with the pillow; and the A/D conversion module 20 can be placed on the bedside or other convenient places. When the subject is sleeping, the pulse signal can be detected by this system, and the processor 30 can continuously process the data during the sleep period of the subject, complete the subsequent analysis, processing and optimization work, and realize the optimization of the R-R interval during sleep . Specifically, the working process and working principle of this system to realize the optimization of the R-R interval during sleep can be understood in combination with the method for measuring and optimizing the pulse R-R interval during sleep of the present invention, so it will not be repeated here.

In other modified embodiments, the processor 30 may be replaced by other intelligent terminal devices such as computers, mobile phones, tablet computers, and watches.

Compared with the prior art, the pulse R-R interval measurement and optimization method and system during sleep of the present invention detect the sleep condition of the subject in real time through non-contact and highly portable equipment and methods with high detection accuracy, and It is only necessary to detect the pulse signal and use a variety of algorithms to achieve accurate R-R interval value extraction. After the initial screening of the pulse peak value through the covariance matching of the pulse peak value, the initial R-R interval value is obtained, and then the singular preliminary R-R interval value is obtained through the AR model. The second investigation of the R-R interval value has achieved the effect of the existing R-R interval value collected by PSG, and can meet the needs of users to optimize the R-R interval in daily life such as home or travel, and reduce Time and money consuming.

The present invention is not limited to the above-mentioned embodiments, if the various changes or deformations of the present invention do not depart from the spirit and scope of the present invention, if these changes and deformations belong to the claims of the present invention and the equivalent technical scope, then the present invention is also It is intended that such modifications and variations are included.

Claims (10)

1. a pulse R-R interval measurement and optimization method during sleep, is characterized in that: comprise the following steps: Step 1: Obtain and calculate the accurate pulse rate, locate the pulse peak point, and accurately obtain the pulse waveform; Step 2: Calculate the preliminary R-R interval value according to the pulse waveform; Step 3: Set a time interval, and establish an AR model used to represent the effective peak information in the time interval according to the preliminary R-R interval value in the time interval; Step 4: The AR model calculates the R-R interval value of the N+1th point according to the preliminary R-R interval value of the first N points in the time interval; Step 5: Calculate the difference between the R-R interval value of the N+1th point calculated by the AR model and the actual R-R interval value of the N+1th point in the time interval; judge Whether the difference falls within a threshold range, if yes, record the actual R-R interval value of the N+1 point in the time interval; otherwise, consider the detection failure and discard the Nth point in the time interval +1 point of actual R-R interval value; Step 6: Sequentially extract a plurality of time intervals with the same duration as the time interval at equal time intervals, and repeat steps 3 to 5; until all time intervals are processed, the R-R interval during sleep is optimized. 2. the pulse R-R interval measurement and optimization method during sleep according to claim 1, is characterized in that: in described step 1, specifically comprise the following steps: Step 11: collect sleep signals in real time; Step 12: Process the sleep signal to obtain the pulse rate; Step 13: Perform morphological filtering, peak detection, and covariance matching processing on the pulse rate in sequence to obtain accurate pulse rate data, and generate a pulse waveform according to the pulse rate data. 3. the pulse R-R interval measurement and optimization method during sleep according to claim 2, is characterized in that: described step 11 is to realize the real-time collection to sleep signal by piezoelectric sensor, and this piezoelectric sensor is placed on during collection The underside of the subject's pillow. 4. The pulse R-R interval measurement and optimization method during sleep according to claim 3 is characterized in that: the type of the sleep signal collected by the piezoelectric sensor in the step 11 is an analog signal type, and the analog signal is sent to an A/D conversion module, which is converted into a sleep signal of digital signal type by the A/D conversion module. 5. the pulse R-R interval measurement and optimization method during sleep according to claim 1, is characterized in that: in described step 2, specifically comprise the following steps: Step 21: Extract a segment of the pulse signal from the pulse waveform, and create a pulse waveform pulse in the segment of the pulse signal; Step 22: moving and matching the pulse waveform pulse with the pulse signal to obtain a set of covariance matching correlation functions; Step 23: Establishing a moving window with a period of 6s in the correlation function, and moving the moving window with a period of 1s in the correlation function, and selecting the maximum value point in the moving window after each movement, And all the selected maximum value points are used as the pulse fluctuation pulse in the pulse signal; Step 24: According to the characteristic that the heart rate does not have a sudden change in a short period of time, judge the pulse period of the pulse fluctuation pulse formed in step 23, remove the abnormal peak value, and obtain effective peak value data; Step 25: mark the obtained effective peak data as the pulse peak point in the correlation function, and calculate the preliminary R-R interval value. 6. The method for measuring and optimizing the pulse R-R interval during sleep according to claim 1, characterized in that: in the step 3, a time interval of 6s is set, and according to 4 to 6 pulses in the time interval The preliminary R-R interval value Xi is used to establish an AR model used to represent the effective peak information in this time interval, where the value of i is an integer between 1 and 6, including 1 and 6. 7. The method for measuring and optimizing the pulse RR interval during sleep according to claim 6, characterized in that: in the step 4, the AR model is based on the first to third points in the time interval Preliminary RR interval values X ₁ , X ₂ and X ₃ are estimated to obtain the RR interval value X ₄ of the fourth point, X ₄ = a×X ₃ +a ² ×X ₂ +a ³ ×X ₁ ; where a The value is the real solution to the equation a+a ² +a ³ =1. 8. the pulse R-R interval measurement and optimization method during sleep according to claim 7, is characterized in that: in described step 5, specifically comprise the following steps: Step 51: Calculate the difference between the RR interval value _X4 of the fourth point calculated by the AR model and the actual RR interval value Y4 of the _fourth point within the time interval; Step 52: Determine whether the absolute value |Y ₄ -X ₄ | of the difference is within X ₄ ×20%, if yes, record the actual RR interval value Y ₄ of the fourth point in the time interval ; Otherwise, the detection is regarded as failure, and the actual RR interval value Y ₄ of the fourth point in the time interval is discarded. 9. the pulse R-R interval measurement and optimization method during sleep according to claim 8, is characterized in that: in described step 6, after the 6s time interval in described step 3, take 1s as time interval, Sequentially extract multiple time intervals of 6 s along the positive direction of the time axis, and repeat the steps 3-5; until all time intervals are processed, the optimization of the R-R interval during sleep is realized. 10. A pulse R-R interval measurement and optimization system during sleep, characterized in that: comprise piezoelectric sensors, A/D conversion modules and processors connected in sequence; The piezoelectric sensor is used for real-time acquisition of sleep analog signals during sleep of the subject, and is transmitted to the A/D conversion module; The A/D conversion module is used to convert the sleep analog signal into a sleep digital signal and transmit it to the processor; the processor is used to process the sleep digital signal to obtain a pulse rate waveform, and The rate waveform is processed to obtain the preliminary R-R interval value, and the preliminary R-R interval value is checked for differences to obtain the optimized R-R interval value, so as to realize the optimization of the R-R interval during sleep; and the processor is used to execute the claim The pulse R-R interval measurement and optimization method during sleep described in any one of 1 to 9.