CN110236522A - Health screening method, system and medical equipment based on single-lead electrocardiogram - Google Patents

Abstract

The invention relates to the technical field of medical systems, and discloses a health screening method, system and medical equipment based on a single-lead electrocardiogram. The health screening method based on a single-lead electrocardiogram includes: collecting a user's single-lead electrocardiogram; analyzing the characteristic data of the collected single-lead electrocardiogram central electrical signal pRRx sequence; combining the characteristic data of the analyzed pRRx sequence with the pre-trained The health detection model performs feature matching; and generates the user's health detection data according to the result of the feature matching. The present invention can be based on a portable ECG monitoring device and a health screening method capable of real-time monitoring, and can perform daily monitoring and health screening on users, which is conducive to timely discovering potential serious disease risks of users, and reminding users to go to the hospital for corresponding examinations in time, diagnosis and timely treatment.

Description

Health screening method, system and medical equipment based on single-lead electrocardiogram

technical field

The invention relates to the technical field of medical equipment, in particular to a single-lead electrocardiogram-based health screening method, system and medical equipment.

Background technique

At present, the process of health detection and diagnosis in hospitals is complicated and cumbersome. Usually, it is necessary to queue up for routine inspections of multiple items such as blood drawing and electrocardiogram. The convenience, comfort, work efficiency and user experience of the whole health data detection are not good. And for users who do not have a high frequency of health checkups, it is easy to miss symptoms and miss the best time for prevention and treatment.

Contents of the invention

In view of this, the present invention provides a single-lead electrocardiogram-based health screening method, system, and medical equipment to solve the technical problems of poor convenience, comfort, work efficiency, and user experience in existing health checks.

According to one embodiment of the present invention, a health screening method based on a single-lead electrocardiogram is provided, including: collecting a user's single-lead electrocardiogram; analyzing the characteristic data of the collected single-lead electrocardiogram central electrical signal pRRx sequence; The feature data of the pRRx sequence is matched with the pre-trained health detection model; and the user's health detection data is generated according to the result of the feature matching.

Preferably, the health screening method based on a single-lead electrocardiogram also includes training a health detection model, which further includes: collecting physiological parameters of different users at different times and corresponding single-lead electrocardiograms; analyzing the collected single-lead electrocardiograms The characteristic data of the central electrical signal pRRx sequence; machine learning and training are performed on the collected physiological parameters and the correspondingly analyzed characteristic data of the pRRx sequence through the training model algorithm to generate a model function of the corresponding relationship between the characteristic data of the pRRx sequence and the physiological parameters; and A health detection model is generated based on the generated model function and the collected physiological parameter data.

Preferably, after the user's health monitoring data is generated according to the result of feature matching, it also includes: acquiring the user's health monitoring data at multiple times; analyzing the acquired health monitoring data at multiple times to obtain the user's health trend data ; and obtain the user's health detection report according to the obtained health monitoring data and health trend data.

Preferably, the feature data of the pRRx sequence is one or a combination of linear features, entropy nonlinear features, and fractal dimension nonlinear features, wherein the linear features are the average value, standard deviation, adjacent One or a combination of the root mean square of the sequence difference and the standard deviation of the adjacent sequence difference, the non-linear characteristics of the fractal dimension are pRRx sequence histogram distribution information entropy, pRRx sequence power spectrum histogram distribution information entropy and pRRx sequence One or a combination of the information entropy of the power spectrum full frequency band distribution, the nonlinear feature of the fractal dimension is the fractal dimension calculated by the structure function method, the fractal dimension calculated by the correlation function method, and the fractal dimension calculated by the variation method One or a combination of fractal dimension and fractal dimension calculated by root mean square method.

Preferably, the analysis of the characteristic data of the collected single-lead electrocardiogram central electrical signal pRRx sequence includes: calculating the difference between the collected single-lead electrocardiographic central electrical signal adjacent pRRx sequences greater than the threshold x milliseconds and all adjacent pRRx The ratio of the number of sequences; and by setting different thresholds x correspondingly to obtain the ratio corresponding to each threshold x to form a pRRx sequence.

According to another embodiment of the present invention, a health screening system based on a single-lead electrocardiogram is also provided, including: an electrocardiogram acquisition device for collecting a user's single-lead electrocardiogram; a feature analysis device for analyzing the electrocardiogram The feature data of the single-lead electrocardiogram central electrical signal pRRx sequence collected by the acquisition device; the feature matching device is used to perform characteristic data of the pRRx sequence analyzed by the feature analysis device and the health detection model pre-trained by the health detection training model device Matching; and detection data generating means for generating user's health detection data according to the result of feature matching by the feature matching means.

Preferably, the health data detection system based on a single-lead electrocardiogram also includes a health detection training model device, which further includes: a data collection unit for collecting physiological parameters of different users at different times and corresponding single-lead electrocardiograms; The data analysis unit is used to analyze the characteristic data of the single-lead electrocardiogram central electrical signal pRRx sequence collected by the data collection unit; the model function training unit is used to use the training model algorithm to collect the physiological parameters and the obtained data collection unit. The data analysis unit performs machine learning and training on the characteristic data of the pRRx sequence corresponding to the analysis, so as to generate a model function of the corresponding relationship between the characteristic data of the pRRx sequence and the physiological parameters; and a health detection model generation unit for training the unit based on the model function The generated model function and the physiological parameter data collected by the data collection unit generate a health detection model.

Preferably, the single-lead electrocardiogram-based health data detection system also includes a detection report generation device, which further includes: a data acquisition unit, used to acquire health monitoring data of the user at multiple times; a trend analysis unit, used for Analyzing the health monitoring data obtained by the data acquisition unit at multiple times to obtain the user's health trend data; and a detection report generation unit for analyzing the health monitoring data obtained by the data acquisition unit and the trend analysis unit The health trend data obtains the user's health detection report.

Preferably, the feature data of the pRRx sequence is one or a combination of linear features, entropy nonlinear features, and fractal dimension nonlinear features, wherein the linear features are the average value, standard deviation, adjacent One or a combination of the root mean square of the sequence difference and the standard deviation of the adjacent sequence difference, the non-linear characteristics of the fractal dimension are pRRx sequence histogram distribution information entropy, pRRx sequence power spectrum histogram distribution information entropy and pRRx sequence One or a combination of the information entropy of the power spectrum full frequency band distribution, the nonlinear feature of the fractal dimension is the fractal dimension calculated by the structure function method, the fractal dimension calculated by the correlation function method, and the fractal dimension calculated by the variation method One or a combination of fractal dimension and fractal dimension calculated by root mean square method.

According to still another embodiment of the present invention, there is also provided a medical device, which includes the above-mentioned health screening system based on a single-lead electrocardiogram.

The health screening method, system and medical equipment based on the single-lead electrocardiogram provided by the present invention collect the user's single-lead electrocardiogram; analyze the characteristic data of the collected single-lead electrocardiogram central electrical signal pRRx sequence; the analyzed pRRx sequence The feature data is matched with the pre-trained health detection model; and the user's health detection data is generated according to the result of the feature matching. The present invention can be based on a portable ECG monitoring device and a health screening method capable of real-time monitoring, and can perform daily monitoring and health screening on users, which is conducive to timely discovering potential serious disease risks of users, and reminding users to go to the hospital for corresponding examinations in time, diagnosis and timely treatment.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings that need to be used in the description of the embodiments. Apparently, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings according to these drawings without creative efforts.

FIG. 1 is a schematic flowchart of a health screening method based on a single-lead electrocardiogram in an embodiment of the present invention.

Fig. 2 is a schematic flow chart of analyzing characteristic data of pRRx sequence in one embodiment of the present invention.

Fig. 3 is a schematic flow chart of training a health detection model in an embodiment of the present invention.

Fig. 4 is a schematic flowchart of generating a user's health inspection report in an embodiment of the present invention.

Fig. 5 is a schematic structural diagram of a health data detection system based on a single-lead electrocardiogram in another embodiment of the present invention.

Fig. 6 is a schematic structural diagram of a health detection training model device in another embodiment of the present invention.

Fig. 7 is a schematic structural diagram of a detection report generating device in another embodiment of the present invention.

Fig. 8 is a schematic structural diagram of a medical device in another embodiment of the present invention.

Detailed ways

The technical solutions of the present invention will be further described in more detail in conjunction with the accompanying drawings and specific embodiments. Apparently, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the description of the present invention, it should be understood that the terms "first", "second" and so on are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance. In the description of the present invention, it should be noted that unless otherwise specified and limited, the terms "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral Ground connection; it can be mechanical connection or electrical connection; it can be direct connection or indirect connection through an intermediary. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention in combination with specific situations. In addition, in the description of the present invention, unless otherwise specified, "plurality" means two or more.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent modules, segments or portions of code comprising one or more executable instructions for implementing specific logical functions or steps of the process , and the scope of preferred embodiments of the invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including substantially concurrently or in reverse order depending on the functions involved, which shall It is understood by those skilled in the art to which the embodiments of the present invention pertain.

Fig. 1 is a schematic flowchart of a health data detection method based on a single-lead electrocardiogram in an embodiment of the present invention. As shown in the figure, the health screening method based on the single-lead electrocardiogram includes:

Step S101: collecting a single-lead electrocardiogram of a user.

In this embodiment, when it is necessary to detect the user's health data, a general-purpose single-lead ECG signal acquisition device can be used to collect the user's single-lead ECG. Check, improving the convenience, comfort, work efficiency and user experience of health check.

Step S102: analyzing the characteristic data of the collected single-lead ECG central electrical signal pRRx sequence.

In this embodiment, the characteristic data of the collected single-lead ECG central electrical signal pRRx sequence are further analyzed. Referring to Fig. 2, the characteristic data of the single-lead electrocardiogram central electrical signal pRRx sequence of described analysis collection, include:

Step S201: Calculate the ratio of the number of adjacent pRRx sequences whose central electrical signal of the collected single-lead ECG is greater than the threshold x milliseconds to the number of all adjacent pRRx sequences.

Step S202: By setting thresholds x with different values correspondingly, obtaining the ratio corresponding to each threshold x to form a pRRx sequence.

In this embodiment, first calculate the ratio of the number of adjacent pRRx sequences whose difference between the collected single-lead ECG central electrical signal is greater than the threshold x milliseconds to the number of all adjacent pRRx sequences, and then obtain corresponding values by setting different thresholds x The ratio corresponding to each threshold x forms the pRRx sequence, and its calculation formula is:

In this embodiment, the feature data of the pRRx sequence is one or a combination of linear features, entropy nonlinear features, and fractal dimension nonlinear features.

The linear feature is one or a combination of the average value AVRR, the standard deviation SDRR of the pRRx sequence, the root mean square rMSSD of the difference between adjacent sequences, and the standard deviation SDSD of the difference between adjacent sequences.

The non-linear feature of fractal dimension is one or a combination of information entropy of pRRx sequence histogram distribution, pRRx sequence power spectrum histogram distribution information entropy and pRRx sequence power spectrum full frequency band distribution information entropy.

For the random variable set A of the probability distribution function p(x), the definition of entropy is shown in formula (2):

H(A)＝-∑p _A (x)logp _A (x) (2)

Available features include:

(1) pRRx sequence histogram distribution information entropy S _dh is the numerical distribution information entropy of pRRx sequence;

(2) The pRRx sequence power spectrum histogram distribution information entropy S _ph is to perform discrete Fourier transform on the pRRx sequence to obtain the power spectrum, and then calculate its information entropy according to the numerical distribution of the power spectrum sequence;

(3) pRRx sequence power spectrum full frequency distribution information entropy S _pf is the power spectrum obtained by discrete Fourier transform of pRRx sequence, in the whole frequency band [f _s /N, f _s /2] (the sampling frequency of the signal is f _s , the number of sampling points is N), inserting i-1 sub-points f ₁ , f ₂ , ..., f _m-1 to divide the whole frequency band into i sub-frequency bands. Taking the sum of the power densities in each frequency band as the power density of the frequency band, m power densities are obtained. Normalize the i power density to obtain the probability p _i of each frequency band, then ∑ _i p _i =1, and the corresponding entropy of the entire frequency band of the power spectrum is shown in formula (3):

In this embodiment, the non-linear feature of the fractal dimension is the fractal dimension calculated by the structure function method, the fractal dimension calculated by the correlation function method, the fractal dimension calculated by the variation method and the root mean square method. One or a combination of the resulting fractal dimensions.

In this embodiment, the pRRx sequence of each segment of ECG is analyzed nonlinearly, and the following four fractal dimension calculation and analysis methods can also be used to obtain the following characteristic indicators:

(1) The fractal dimension D _sf calculated by the structural function method, wherein the structural function method refers to defining the incremental variance as a structural function for a given sequence z(x), and its relationship is:

For several scales τ, calculate the corresponding S(τ) for the discrete values of the sequence z(x), then draw the function curve of logS(τ)-logτ, and perform linear fitting in the scale-free area to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _sf and the slope α is shown in formula (5):

(2) The fractal dimension D _cf calculated by the correlation function method, where the correlation function method means that for a given sequence z(x), the correlation function C(τ) is defined as shown in formula (6):

C(τ)=AVE(z(x+τ)*z(x)), τ=1, 2, 3, ..., N-1 (6)

Among them, AVE(·) represents the average, and τ represents the distance between two points. At this time, the correlation function is a power type, and since there is no characteristic length, the distribution is a fractal, with C(τ)ατ ^-α . At this time, draw the function curve of logC(τ)-logτ, and perform linear fitting in the scale-free area to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _cf and the slope α is shown in formula (7):

D _cf =2-α (7)

(3) The fractal dimension D _vm calculated by the variation method, wherein, the variation method covers the fractal curve with a rectangular frame with a width of τ connected end to end, so that the maximum value and minimum value of the curve in the i-th frame The difference is H(i), which is the height of the rectangle. Multiply the height and width of all rectangles to get the total area S(τ). Change the size of τ to get a series of S(τ). As shown in formula (8):

Draw the function curve of log N(τ)-logτ, and perform linear fitting in the scale-free region to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _vm and slope α is shown in formula (7).

(4) The fractal dimension D _rms calculated by the root mean square method, wherein the root mean square method covers the fractal curve with a rectangular frame with a width τ connected end to end, so that the maximum value of the curve in the i-th frame and The minimum difference is H(i), which is the height of the rectangle. Compute the root mean square value S(τ) of the heights of these rectangles. Change the size of τ to get a series of S(τ). Draw the function curve of log S(τ)-log τ, and perform linear fitting in the scale-free region to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _rms and the slope α is shown in formula (7).

Step S103: performing feature matching on the analyzed feature data of the pRRx sequence and the pre-trained health detection model.

Referring to Figure 3, before feature matching, the health detection model needs to be pre-trained. The training health detection model includes:

Step S301: collecting physiological parameters and corresponding single-lead electrocardiograms of different users at different times.

Step S302: analyzing the characteristic data of the collected single-lead ECG central electrical signal pRRx sequence.

Step S303: Carry out machine learning and training on the collected physiological parameters and correspondingly analyzed characteristic data of the pRRx sequence through the training model algorithm to generate a model function of the corresponding relationship between the characteristic data of the pRRx sequence and the physiological parameters.

Step S304: Generate a health detection model based on the generated model function and the collected physiological parameter data.

In this embodiment, first collect the physiological parameters of different users at different times and the corresponding single-lead electrocardiogram, then analyze the characteristic data of the collected single-lead electrocardiogram center electrical signal pRRx sequence, and then use the training model algorithm to analyze the collected physiological parameters And machine learning and training are performed on the characteristic data of the corresponding analyzed pRRx sequence to generate a model function of the corresponding relationship between the characteristic data of the pRRx sequence and physiological parameters, and finally generate a health detection model based on the generated model function and the collected physiological parameter data.

Among them, the physiological parameters include but are not limited to: age, gender, medical history, blood sugar, body mass index, blood pressure, urine routine, blood potassium, hemoglobin, blood routine, blood creatinine, blood lipids, uric acid, hemodynamic monitoring results, ultrasound Cardiogram or carotid ultrasound, urine protein, X-ray film, electrocardiogram, fundus, smoking history, intracardiac electrophysiological examination results, coronary angiography results, disease type, stage, number and severity of complications, etc.

After analyzing and obtaining the user's characteristic data and the health detection model, further perform feature matching on the analyzed and obtained characteristic data and the pre-trained health detection model. Specifically, all the feature data acquired by analysis are matched item by item with all feature templates in the pre-trained health detection model. When the similarity between specific feature data and one of the feature templates is found to exceed the preset threshold, the The feature matching is successful, otherwise it is judged that the feature matching between the two fails.

Step S104: Generate user's health detection data according to the result of feature matching.

In this embodiment, when the characteristic data acquired by analysis matches the specific characteristic template in the pre-trained health detection model, according to the model function of the corresponding relationship between the characteristic data of the central electrical signal pRRx sequence of the health detection model and the physiological parameters, it can be Output the physiological parameter data corresponding to the specific feature template to generate the detection data of the current user's health, which can conveniently and quickly evaluate the user's health status based on the health detection data, screen the situation with disease risk, and remind the user Seek medical diagnosis in time and treat at the best time.

Referring to FIG. 4 , in some other embodiments, on the basis of the above embodiments, a user's health detection report may be further generated. The user's health detection report generated includes:

Step S401: Obtain the user's health monitoring data at multiple times.

Step S402: Analyze the acquired health monitoring data at multiple times to obtain the user's health trend data.

Step S403: Obtain the user's health detection report according to the obtained health monitoring data and health trend data.

In this embodiment, single-lead ECG detection can be performed on the same user at multiple times, the user's health monitoring data at multiple times can be obtained, and the health monitoring data obtained at multiple times can be analyzed to obtain the user's health trend data. And according to the obtained health monitoring data and health trend data, the user's health detection report can be obtained. Based on the portable ECG monitoring equipment and health screening method that can be monitored in real time, it can conduct daily monitoring and health screening for users, which is conducive to timely Discover the user's potential risk of serious illness, and remind the user to go to the hospital for corresponding examination, diagnosis and timely treatment.

Fig. 5 is a schematic structural diagram of a health data detection system based on a single-lead electrocardiogram in another embodiment of the present invention. Based on the above method embodiments, the single-lead ECG-based health data detection system 100 of this embodiment includes an ECG acquisition device 10 , a feature analysis device 20 , a feature matching device 30 and a detection data generation device 40 .

In this embodiment, when it is necessary to detect the user's health data, the user's single-lead ECG can be collected by the electrocardiogram acquisition device 10, such as a general-purpose single-lead ECG signal acquisition device, and the acquisition process is simple and non-invasive. There is no need to queue up for routine inspections of various items, which improves the convenience, comfort, work efficiency and user experience of health inspections.

In this embodiment, the characteristic analysis device 20 further analyzes the characteristic data of the central electrical signal pRRx sequence of the single-lead electrocardiogram collected by the electrocardiogram collection device 10 . The feature analysis device 20 includes first calculating the ratio of the number of adjacent pRRx sequences in which the difference between the collected single-lead electrocardiogram central electrical signal is greater than the threshold x milliseconds to the number of all adjacent pRRx sequences, and then by setting different thresholds x to correspond to The ratio corresponding to each threshold x is obtained to form a pRRx sequence, and its calculation formula is:

In this embodiment, the feature data of the pRRx sequence is one or a combination of linear features, entropy nonlinear features, and fractal dimension nonlinear features.

The linear feature is one or a combination of the average value AVRR, the standard deviation SDRR of the pRRx sequence, the root mean square rMSSD of the difference between adjacent sequences, and the standard deviation SDSD of the difference between adjacent sequences.

The non-linear feature of fractal dimension is one or a combination of information entropy of pRRx sequence histogram distribution, pRRx sequence power spectrum histogram distribution information entropy and pRRx sequence power spectrum full frequency band distribution information entropy.

For the random variable set A of the probability distribution function p(x), the definition of entropy is shown in formula (2):

H(A)＝-∑p _A (x)log p _A (x) (2)

Available features include:

(4) pRRx sequence histogram distribution information entropy S _dh is the numerical distribution information entropy of pRRx sequence;

(5) The pRRx sequence power spectrum histogram distribution information entropy S _ph is to perform discrete Fourier transform on the pRRx sequence to obtain the power spectrum, and then calculate its information entropy according to the numerical distribution of the power spectrum sequence;

(6) pRRx sequence power spectrum full frequency distribution information entropy S _pf is the power spectrum obtained by discrete Fourier transform of pRRx sequence, in the whole frequency band [f _s /N, f _s /2] (the sampling frequency of the signal is f _s , the number of sampling points is N), inserting i-1 sub-points f ₁ , f ₂ , ..., f _m-1 to divide the whole frequency band into i sub-frequency bands. Taking the sum of the power densities in each frequency band as the power density of the frequency band, m power densities are obtained. Normalize the i power density to obtain the probability p _i of each frequency band, then ∑ _i p _i =1, and the corresponding entropy of the entire frequency band of the power spectrum is shown in formula (3):

In this embodiment, the non-linear feature of the fractal dimension is the fractal dimension calculated by the structure function method, the fractal dimension calculated by the correlation function method, the fractal dimension calculated by the variation method and the root mean square method. One or a combination of the resulting fractal dimensions.

In this embodiment, the pRRx sequence of each segment of ECG is analyzed nonlinearly, and the following four fractal dimension calculation and analysis methods can also be used to obtain the following characteristic indicators:

(5) The fractal dimension D _sf calculated by the structural function method, wherein the structural function method refers to defining the incremental variance as a structural function for a given sequence z(x), and its relationship is:

For several scales τ, the corresponding S(τ) is calculated for the discrete values of the sequence z(x), and then the function curve of log S(τ)-logτ is drawn, and linear fitting is carried out in the scale-free region to obtain The slope α, the conversion relationship between the corresponding fractal dimension D _sf and the slope α is shown in formula (5):

(6) The fractal dimension D _cf calculated by the correlation function method, where the correlation function method means that for a given sequence z(x), the correlation function C(τ) is defined as shown in formula (6):

C(τ)=AVE(z(x+τ)*z(x)), τ=1, 2, 3, ..., N-1 (6)

Among them, AVE(·) represents the average, and τ represents the distance between two points. At this time, the correlation function is a power type, and since there is no characteristic length, the distribution is a fractal, with C(τ)ατ ^-α . At this time, draw the function curve of log C(τ)-logτ, and perform linear fitting in the scale-free area to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _cf and the slope α is shown in formula (7) :

D _cf =2-α (7)

(7) The fractal dimension D _vm calculated by the variation method, wherein, the variation method covers the fractal curve with a rectangular frame with a width of τ connected end to end, so that the maximum value and minimum value of the curve in the i-th frame The difference is H(i), which is the height of the rectangle. Multiply the height and width of all rectangles to get the total area S(τ). Change the size of τ to get a series of S(τ). As shown in formula (8):

Draw the function curve of log N(τ)-logτ, and perform linear fitting in the scale-free region to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _vm and slope α is shown in formula (7).

(8) The fractal dimension D _rms calculated by the root mean square method, wherein the root mean square method covers the fractal curve with a rectangular frame with a width of τ connected end to end, so that the maximum value of the curve in the i-th frame and The minimum difference is H(i), which is the height of the rectangle. Compute the root mean square value S(τ) of the heights of these rectangles. Change the size of τ to get a series of S(τ). Draw the function curve of log S(τ)-logτ, and perform linear fitting in the scale-free region to obtain the slope α, then the conversion relationship between the corresponding fractal dimension D _rms and the slope α is shown in formula (7).

Referring to FIG. 6 , before the feature matching device 30 performs feature matching, the health detection model needs to be pre-trained by the health detection training model device 50 . The health detection training model device 50 includes a data collection unit 501 , a data analysis unit 502 , a model function training unit 503 and a health detection model generation unit 504 .

In this embodiment, firstly, the data collection unit 501 collects physiological parameters and corresponding single-lead ECGs of different users at different times, and then the data analysis unit 502 analyzes the center of the single-lead ECGs collected by the data collection unit 501. The characteristic data of the electrical signal pRRx sequence, the model function training unit 503 performs machine learning and analysis on the physiological parameters collected by the data acquisition unit 501 and the characteristic data of the pRRx sequence correspondingly analyzed by the data analysis unit 502 through the training model algorithm Training to generate the model function of the corresponding relationship between the characteristic data of the pRRx sequence and the physiological parameters, and finally the health detection model generation unit 504 is based on the model function generated by the model function training unit 503 and the physiological parameters collected by the data acquisition unit 501 The data generates a health detection model.

Among them, the physiological parameters include but are not limited to: age, gender, medical history, blood sugar, body mass index, blood pressure, urine routine, blood potassium, hemoglobin, blood routine, blood creatinine, blood lipids, uric acid, hemodynamic monitoring results, ultrasound Cardiogram or carotid ultrasound, urine protein, X-ray film, electrocardiogram, fundus, smoking history, intracardiac electrophysiological examination results, coronary angiography results, disease type, stage, number and severity of complications, etc.

After the feature analysis device 20 analyzes and acquires the user's feature data and the health detection model, the feature matching device 30 further combines the feature data acquired by the feature analysis device 20 with the health test model pre-trained by the health detection training model device 50. The detection model performs feature matching. Specifically, the feature matching device 30 performs similarity matching on all the feature data acquired by the feature analysis device 20 and all feature templates in the health detection model pre-trained by the health detection training model device 50 item by item. When it is found that the similarity between the specific feature data and one of the feature templates exceeds the preset threshold, it is judged that the feature matching of the two is successful, otherwise it is judged that the feature matching of the two fails.

In this embodiment, when the feature matching device 30 matches the feature data acquired by the feature analysis device 20 with the specific feature template in the health detection model pre-trained by the health detection training model device 50, the detection The data generation device 40 can output the physiological parameter data corresponding to the specific characteristic template according to the model function of the corresponding relationship between the characteristic data of the electric signal pRRx sequence in the health detection model center and the physiological parameter to generate the detection data of the current user's physical health, and improve It improves the convenience, comfort, work efficiency and user experience of health screening.

In some embodiments, after the detection data generating device 40 generates the health detection data according to the result of feature matching, when the generated detection data exceeds the preset warning range, it can further use text, voice or jump out of the alarm window, etc. Ways to remind detection data, see Figure 7, in some other embodiments, on the basis of the above embodiments, a detection report generation device 70 may be further included, which includes a data acquisition unit 701, a trend analysis unit 702 and a detection report generation unit 703 .

In this embodiment, single-lead ECG detection can be performed at multiple times for the same user, the data acquisition unit 701 acquires the health monitoring data of the user at multiple times, and the trend analysis unit 702 analyzes the data acquisition unit 701 obtains the health monitoring data of multiple times to obtain the user’s health trend data, and the detection report generation unit 703 obtains the user’s health detection report according to the health monitoring data and the health trend data, and can perform daily monitoring and health screening for the user. It is helpful to timely discover the potential risk of serious illness of the user, and remind the user to go to the hospital for corresponding examination, diagnosis and timely treatment in time.

Fig. 8 is a schematic structural diagram of a medical device in another embodiment of the present invention. As shown in the figure, another embodiment of the present invention also provides a medical device 200, the medical device 200 includes the health data detection system 100 based on the single-lead electrocardiogram in the above embodiment, which can perform daily monitoring on the user and make Health screening is conducive to timely detection of potential serious disease risks of users, and reminds users to go to the hospital for corresponding examination, diagnosis and timely treatment.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the embodiments described above, various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques known in the art: Discrete logic circuits, ASICs with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

In the description of this specification, reference to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" means that specific features described in connection with the embodiment or example, A structure, material or characteristic is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications, substitutions and modifications can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the invention is defined by the claims and their equivalents.

Claims (10)

1. a kind of human health screening method based on single lead electrocardiogram characterized by comprising

Acquire the single lead electrocardiogram of user；

Analyze the characteristic of the single lead electrocardiogram center telecommunications pRRx sequence of acquisition；

The characteristic of the pRRx sequence of analysis and health detection model trained in advance are subjected to characteristic matching；And

The health detection data of user are generated according to the result of characteristic matching.

2. the human health screening method according to claim 1 based on single lead electrocardiogram, which is characterized in that further include training Health detection model, further comprising:

Acquire the physiological parameter and corresponding single lead electrocardiogram of different user different time；

Analyze the characteristic of the single lead electrocardiogram center telecommunications pRRx sequence of acquisition；

Machine is carried out by characteristic of the training pattern algorithm to the physiological parameter of acquisition and the pRRx sequence of correspondence analysis Study and training, to generate the characteristic of pRRx sequence and the pattern function of physiological parameter corresponding relationship；And

The physiological parameter data of pattern function and acquisition based on generation generates health detection model.

3. the health data detection method according to claim 1 or 2 based on single lead electrocardiogram, which is characterized in that After the health detection data for generating user according to the result of characteristic matching, further includes:

User is obtained in the health monitoring data of multiple times；

The health monitoring data of the multiple times obtained are analyzed to obtain the healthy trend data of user；And

It is reported according to the health monitoring data of acquisition and the health detection of healthy trend data acquisition user.

4. the health data detection method according to claim 1 or 2 based on single lead electrocardiogram, which is characterized in that institute The characteristic for stating pRRx sequence is one of linear character and entropy nonlinear characteristic, fractal dimension nonlinear characteristic or group It closes, wherein the linear character is poor for average value, standard deviation, the root mean square of flanking sequence difference, the flanking sequence of pRRx sequence One of standard deviation of value or combination, the fractal dimension nonlinear characteristic are pRRx sequence histogram distributed intelligence entropy, pRRx Sequence power is composed one of histogram distributed intelligence entropy and pRRx sequence power spectrum full frequency band distributed intelligence entropy or is combined, described point Shape dimension nonlinear characteristic calculates resulting fractal dimension for structure function method, correlation function algorithm calculates resulting fractal dimension, Variate-difference method calculates resulting fractal dimension and mean square root method calculates one of resulting fractal dimension or combination.

5. the human health screening method according to claim 1 based on single lead electrocardiogram, which is characterized in that the analysis is adopted The characteristic of the single lead electrocardiogram center telecommunications pRRx sequence of collection, comprising:

The difference for calculating the adjacent pRRx sequence of single lead electrocardiogram center telecommunications number of acquisition is greater than threshold value x milliseconds of quantity and whole The ratio of the quantity of adjacent pRRx sequence；And

PRRx sequence is formed by the corresponding corresponding ratio of each threshold value x that obtains of the different threshold value x of setting value.

6. a kind of human health screening system based on single lead electrocardiogram characterized by comprising

Electrocardiogram acquisition device, for acquiring the single lead electrocardiogram of user；

Feature analyzing apparatus, for analyzing the single lead electrocardiogram center telecommunications pRRx sequence of the electrocardiogram acquisition device acquisition The characteristic of column；

Characteristic matching device, characteristic and the health detection training of the pRRx sequence for analyzing the feature analyzing apparatus The health detection model that model equipment is trained in advance carries out characteristic matching；And

Detection data generating means, for generating the health detection of user according to the matched result of the characteristic matching device characteristic Data.

7. the human health screening system according to claim 6 based on single lead electrocardiogram, which is characterized in that further include health Training pattern device is detected, further comprising:

Data acquisition unit, for acquiring the physiological parameter and corresponding single lead electrocardiogram of different user different time；

Data analysis unit, for analyzing the single lead electrocardiogram center telecommunications pRRx sequence of the data acquisition unit acquisition Characteristic；

Pattern function training unit, physiological parameter for being acquired by training pattern algorithm to the data acquisition unit and The characteristic of the pRRx sequence of the data analysis unit correspondence analysis carries out machine learning and training, to generate pRRx sequence Characteristic and physiological parameter corresponding relationship pattern function；And

Health detection model generation unit, pattern function and the data for being generated based on the pattern function training unit The physiological parameter data of acquisition unit acquisition generates health detection model.

8. the human health screening system according to claim 6 or 7 based on single lead electrocardiogram, which is characterized in that further include Examining report generating means, further comprising:

Data capture unit, for obtaining user in the health monitoring data of multiple times；

Trend analysis unit, for analyzing the health monitoring data for multiple times that the data capture unit obtains to obtain use The healthy trend data at family；And

Examining report generation unit, health monitoring data and the trend analysis for being obtained according to the data capture unit The health detection report of the healthy trend data acquisition user of unit analysis.

9. the health data detection system according to claim 6 or 7 based on single lead electrocardiogram, which is characterized in that institute The characteristic for stating pRRx sequence is one of linear character and entropy nonlinear characteristic, fractal dimension nonlinear characteristic or group It closes, wherein the linear character is poor for average value, standard deviation, the root mean square of flanking sequence difference, the flanking sequence of pRRx sequence One of standard deviation of value or combination, the fractal dimension nonlinear characteristic are pRRx sequence histogram distributed intelligence entropy, pRRx Sequence power is composed one of histogram distributed intelligence entropy and pRRx sequence power spectrum full frequency band distributed intelligence entropy or is combined, described point Shape dimension nonlinear characteristic calculates resulting fractal dimension for structure function method, correlation function algorithm calculates resulting fractal dimension, Variate-difference method calculates resulting fractal dimension and mean square root method calculates one of resulting fractal dimension or combination.

10. a kind of Medical Devices, which is characterized in that the Medical Devices include being based on as claim 6 to 9 is described in any item The human health screening system of single lead electrocardiogram.