Method of monitoring heart rhythm and related devices

Abstract

The application discloses a heart rhythm monitoring method and related device, the method comprises the following steps: receiving a first preset period of physiological signals collected by an optical sensor, and extracting the heartbeat cycle in the physiological signals to obtain a first heartbeat cycle sequence; determining a first entropy value of the first heartbeat cycle sequence; determining a first admission threshold of the first entropy value according to a plurality of first reference thresholds determined at different time spans; determining the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain a first cycle difference sequence; determining a second entropy value of the first cycle difference sequence; determining a second admission threshold of the second entropy value according to a plurality of second reference thresholds determined at different time spans and a heart rate variability coefficient; obtaining a heart rhythm monitoring result according to the first entropy value, the second entropy value, the first admission threshold and the second admission threshold; and controlling the heart rhythm monitoring result to be displayed on a display screen. The application can improve the monitoring accuracy of the heart rhythm.

Description

Technical Field

[0001] This application relates to the field of signal processing technology, and in particular to a method and related apparatus for monitoring heart rhythm. Background Technology

[0002] In recent years, with increased health awareness, more and more people are paying attention to their cardiovascular health. At the same time, the widespread use of smart wearable devices has facilitated personal health monitoring. However, existing heart rate monitoring technologies lack sufficient accuracy and cannot comprehensively and accurately assess abnormalities in heart rhythm. Summary of the Invention

[0003] This application provides a method and related apparatus for monitoring heart rhythm, so as to improve the accuracy of heart rhythm monitoring.

[0004] In a first aspect, embodiments of this application provide a method for monitoring heart rhythm, applied to a processor of a smart wearable device, the smart wearable device further including an optical sensor and a display screen, the method comprising:

[0005] The system receives the physiological signal to be identified during a first preset time period collected by the optical sensor, and extracts the heartbeat cycle from the physiological signal to be identified to obtain the first heartbeat cycle sequence.

[0006] Determine a first entropy value for the first heartbeat cycle sequence; and determine a first admission threshold for the first entropy value based on a plurality of first reference thresholds determined for different time spans;

[0007] Determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain the first cycle difference sequence;

[0008] Determine a second entropy value for the first period difference sequence; and determine a second admission threshold for the second entropy value based on multiple second reference thresholds determined by different time spans and heart rate variability coefficients, wherein the heart rate variability coefficient is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans;

[0009] The cardiac rhythm monitoring results are obtained based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold;

[0010] The heart rhythm monitoring results are displayed on the display screen.

[0011] Secondly, embodiments of this application provide a heart rhythm monitoring device applied to a processor of a smart wearable device, the smart wearable device further including an optical sensor and a display screen, the device comprising:

[0012] The extraction unit is used to receive the physiological signal to be identified during a first preset time period collected by the optical sensor, and extract the heartbeat cycle from the physiological signal to be identified to obtain the first heartbeat cycle sequence.

[0013] The first determining unit is configured to determine a first entropy value of the first heartbeat cycle sequence; and to determine a first admission threshold for the first entropy value based on a plurality of first reference thresholds determined according to different time spans.

[0014] The second determining unit is used to determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain the first cycle difference sequence.

[0015] The third determining unit is used to determine the second entropy value of the first period difference sequence; and to determine the second admission threshold of the second entropy value based on a plurality of second reference thresholds determined by different time spans and heart rate variability coefficients, wherein the heart rate variability coefficient is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans;

[0016] The monitoring unit is used to obtain the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold;

[0017] The display unit is used to control the display of the heart rhythm monitoring results on the display screen.

[0018] Thirdly, embodiments of this application provide an electronic device including a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code and performs the steps of the method as described in the first aspect.

[0019] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable program code, the executable program code including execution instructions for performing the steps of the method as described in the first aspect.

[0020] Fifthly, embodiments of this application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in the first aspect of embodiments of this application. The computer program product may be a software installation package.

[0021] As can be seen, in this embodiment, the process first receives the physiological signal to be identified during a first preset time period collected by the optical sensor, and extracts the heartbeat cycle from the physiological signal to obtain a first heartbeat cycle sequence; then, a first entropy value of the first heartbeat cycle sequence is determined; and a first admission threshold of the first entropy value is determined based on multiple first reference thresholds determined by different time spans; the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence is determined to obtain a first cycle difference sequence; then, a second entropy value of the first cycle difference sequence is determined; and a second admission threshold of the second entropy value is determined based on multiple second reference thresholds determined by different time spans and the coefficient of variation of heart rate, wherein the coefficient of variation of heart rate is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans; and a heart rhythm monitoring result is obtained based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold; finally, the heart rhythm monitoring result is controlled to be displayed on the display screen.

[0022] This application monitors heart rhythm using multimodal data, including the first entropy value of the heartbeat cycle sequence and the second entropy value of the cycle difference sequence. This can comprehensively reflect the changing characteristics of heart rhythm, meet the high-precision requirements of monitoring, and improve the accuracy of monitoring results. At the same time, when assessing the regularity of heart rhythm, multiple thresholds that are automatically adjusted according to different time spans and / or considering the coefficient of variation of heart rate are added, which is conducive to flexibly adapting to the dynamic changes of physiological signals, thereby further improving the accuracy of monitoring results. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a system architecture diagram of a heart rhythm monitoring system provided in an embodiment of this application;

[0025] Figure 2 This is a schematic flowchart of a method for monitoring heart rhythm provided in an embodiment of this application;

[0026] Figure 3 This is a schematic diagram of a classic signal provided in an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of a signal with slight noise provided in an embodiment of this application;

[0028] Figure 5 This is a schematic diagram illustrating poor signal quality provided in an embodiment of this application;

[0029] Figure 6 This is a histogram of the fourth entropy value provided in an embodiment of this application;

[0030] Figure 7 This is a histogram of a first entropy value provided in an embodiment of this application;

[0031] Figure 8 This is a histogram of a second entropy value provided in an embodiment of this application;

[0032] Figure 9 This is a schematic diagram of a normal heart rhythm provided in an embodiment of this application;

[0033] Figure 10 This is a schematic diagram of a mildly irregular heart rhythm provided in an embodiment of this application;

[0034] Figure 11 This is a schematic diagram of a severe irregular heart rhythm provided in an embodiment of this application;

[0035] Figure 12 This is a schematic diagram of a display screen provided in an embodiment of this application;

[0036] Figure 13 This is a block diagram of the functional units of a heart rhythm monitoring device provided in an embodiment of this application;

[0037] Figure 14 This is a block diagram of the functional units of another cardiac rhythm monitoring device provided in this application embodiment;

[0038] Figure 15 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application. Detailed Implementation

[0039] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0040] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0041] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0042] With the accelerating pace of modern life and increasing work pressure, cardiovascular disease has become a major public health problem worldwide. According to the World Health Organization, cardiovascular disease causes approximately 17.9 million deaths globally each year, accounting for 32% of all deaths. In China, the incidence of cardiovascular disease is also rising annually, becoming the leading cause of death among residents. Particularly among middle-aged and elderly people, the incidence of heart diseases such as arrhythmia is high, seriously affecting their quality of life and lifespan. However, existing heart rate monitoring technologies mainly rely on single data points, resulting in insufficient monitoring accuracy and an inability to comprehensively and accurately assess abnormalities in heart rhythm.

[0043] To address the aforementioned problems, this application provides a method and related apparatus for monitoring heart rhythm. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0044] Please see Figure 1 , Figure 1 This is a system architecture diagram of a heart rhythm monitoring system provided in an embodiment of this application. Figure 1 As shown, the heart rhythm monitoring system 100 includes a processor 101, a display 102, and an optical sensor 103. The processor 101, the display 102, and the optical sensor 103 are interconnected. The optical sensor 103 is used to collect continuous physiological signals for a certain duration and transmit them to the processor 101. The optical sensor 103 is a non-invasive monitoring device that can continuously collect data for a long time, including but not limited to a photoplethysmography (PPG) recorder.

[0045] The processor 101 is the control core of the system. Based on user operations or preset programs, it sends instructions to the display 102 and the optical sensor 103, such as controlling the optical sensor 103 to collect and transmit physiological signals, and controlling the display 102 to display monitoring results. Specifically, the processor 101 receives physiological signals from the optical sensor 103 and performs operations such as signal quality assessment, heart rate cycle change assessment, and dynamic change characteristics assessment of the heart rate cycle based on these physiological signals, thereby generating heart rhythm monitoring results. Simultaneously, it sends the generated heart rhythm monitoring results to the display 102, enabling the display 102 to receive and display the results.

[0046] Specifically, the display 102 is used to present the detection results transmitted from the processor 101 in an intuitive image form and / or text description, providing users with high-precision monitoring results.

[0047] Based on this, this application provides a method and related device for monitoring heart rhythm, which will be described in detail below with reference to the accompanying drawings.

[0048] Please see Figure 2 , Figure 2 This is a flowchart illustrating a method for monitoring heart rhythm provided in an embodiment of this application, as shown below. Figure 2 As shown, the method includes the following steps:

[0049] S210, receive the physiological signal to be identified during the first preset time period collected by the optical sensor, and extract the heartbeat cycle from the physiological signal to be identified to obtain the first heartbeat cycle sequence.

[0050] This method uses optical sensors, such as photoplethysmography (PPG), to collect continuous physiological signals over a preset time period, which can be 30 seconds, for example. The use of optical sensors for non-invasive monitoring eliminates the need for electrode patches or wires, making it comfortable to wear and suitable for long-term use.

[0051] Specifically, before extracting the heartbeat cycle from the physiological signal to be identified to obtain the heartbeat cycle sequence, and after receiving the physiological signal to be identified from the optical sensor during a first preset time period, the method further includes: determining a fourth entropy value of the physiological signal to be identified; determining a third median and a third standard deviation of multiple fifth entropy values, wherein a single fifth entropy value is used to indicate the entropy value of a single first physiological signal within the time period corresponding to the time span n; determining a third reference threshold under the time span n based on a sixth adjustment coefficient, the third median, and the third standard deviation; obtaining a third threshold and a fourth threshold, wherein the third threshold is a threshold for the entropy value of the second physiological signal actually used in the first time period, and the fourth threshold is a threshold for the entropy value of the third physiological signal used for reference adjustment in the second time period; fusing the third threshold and the fourth threshold to obtain a fourth reference threshold under the time span m; fusing the third reference threshold and the fourth reference threshold to obtain a third admission threshold for the fourth entropy value; and detecting that the fourth entropy value is less than the third admission threshold.

[0052] Upon receiving the physiological signal to be identified, the signal quality of the signal can be evaluated. Specifically, the signal quality is evaluated using the entropy value of the physiological signal. The fourth entropy value can be obtained by calculating the information entropy, sample entropy, or approximate entropy of the physiological signal to be identified.

[0053] For example, the fourth entropy value of the 30-second physiological signal to be identified can be calculated using Shannon entropy (information entropy). Specifically, the 30-second photosensitive physiological signal X[n] is preprocessed to ensure the cleanliness and consistency of the data.

[0054] First, normalization is performed. Preferably, the signal is normalized to a standard range, such as [0, 1] or [-1, 1]. After preprocessing, a signal vector X[n] = {x1, x2, ..., xN} with a length of N is obtained.

[0055] Then, the signal vector X[n] is quantized to convert the continuous signal value into a discrete state. For example, this can be achieved by dividing the signal value into several intervals. The signal value is divided into M intervals, and the width of each interval is Δ. Specifically, the number of intervals is determined first. Commonly used methods for selecting the number of intervals include the Sturges formula, the Scott rule, or the Freedman-Diaconis rule. Preferably, in this embodiment, the Sturges formula is used: M = 1 + log2(N), where N is the length of the signal. Then, the interval width Δ is calculated, and its calculation formula (1) is as follows: Finally, the signal is quantized, mapping each signal value xi to a corresponding interval. For example, if xi falls within the k-th interval, it is quantized to k.

[0056] Furthermore, the interval width can be adaptively adjusted. If the interval width is adjusted, a 7-day adaptive modeling period is required; real-time judgment capability only begins after the 8th day. Specifically, the formula (2) for calculating the adaptive interval width is as follows: Δ adj =mean(Δ day_k ), where Δ day_k Let represent the interval width calculated on day k, where k = 1 to 7. The interval width on day eight is the average of the interval widths of the previous seven days. The interval widths after day eight are dynamically weighted, and the calculation formula (3) is as follows: Δ adj_NEW =Δ adj ×0.9+Δ day_k ×0.1, where Δ adj_NEW Δ is the adaptive interval width for the day. adj The historical adaptive interval width, Δ, was calculated two days ago using formula (3). day_k The historical adaptive interval width calculated one day ago according to formula (1), or Δ day_k The historical adaptive interval width is calculated one day ago using formulas (1) and (2).

[0057] Furthermore, based on the quantized signal values, the frequency of occurrence in each interval, i.e., the probability distribution, is calculated. Specifically, the number of signal values ​​in each interval is first counted, where nk represents the number of signal values ​​in the k-th interval. Then, the probability is calculated: Where N is the total signal length, and pk represents the number of signal values ​​in the k-th interval and the probability of the signal's occurrence. Then, the entropy value H of the physiological signal is calculated:

[0058] Specifically, the fourth entropy value of the physiological signal to be identified is calculated using the entropy calculation method described above. The first entropy value of the heartbeat cycle sequence and the second entropy value of the period difference sequence can also be calculated using the same method.

[0059] The admission threshold corresponding to the fourth entropy value is also adaptively adjusted. Specifically, the median of the fourth entropy value H4 calculated for each physiological signal segment throughout the day is taken, and the calculation formula is as follows:

[0060] H4 day_k =median(H4) i ),

[0061] The median function represents all physiological signal segments H4. i After sorting, take the median value, H4 day_kAs a representative fourth entropy value for day k, for example, the fourth entropy value H4 for day one. day_1 =2.84.

[0062] Next, the standard deviation of the fourth entropy value H4 calculated for each physiological signal segment throughout the day is determined using the following formula:

[0063] H4 std_k =std(H4) i ),

[0064] Here, the std function represents the computation set {H4}. i Standard deviation, H4 std_k As a standard deviation representing the fourth entropy value on day k, for example, the standard deviation H4 of the fourth entropy value on day 1. std_1 =1.02.

[0065] When adaptively adjusting the admission threshold, a 7-day adaptive modeling period is required. Real-time judgment capability is only available after the 8th day. The admission threshold includes the first admission threshold corresponding to the first entropy value, the second admission threshold corresponding to the second entropy value, and the third admission threshold corresponding to the fourth entropy value.

[0066] Specifically, for the third admission threshold corresponding to the fourth entropy value, the calculation formula (4) for its adaptive admission threshold is as follows:

[0067] α4 day_k =H4 day_k -H4 std_k ×0.4,

[0068] Among them, α4 day_k Let be the admission threshold for day k, where 0.4 is an adjustment coefficient, an empirical threshold that can be adjusted according to actual conditions. Furthermore, for day eight, the average admission threshold α4 of the physiological signals from the previous seven days is used. adj As a signal admission standard, its calculation formula (5) is as follows: α4 adj =mean(α4) day_k ), where α4 day_k The threshold value represents the physiological signal calculated on day k, where k = 1 to 7. The threshold value of the fourth entropy value after day eight is dynamically weighted, and its calculation formula (6) is as follows:

[0069] α4 adj_NEW =α4 adj ×0.9+α4 recent ×0.1,

[0070] Among them, α4 adj_NEWThe adaptive admission threshold for the fourth entropy value of the day represents a new dynamically adaptive admission threshold used to determine the quality of the signal. α4 adj The historical adaptive admission threshold calculated two days ago using formula (6) is the actual admission threshold used, α4. recent The historical adaptive admission threshold determined one day prior according to formula (4) is used as the reference adjustment admission threshold. Further, α4 recent It can also be the historical adaptive admission threshold determined one day ago based on calculation formulas (4) and (5); 0.9 is α4. adj The weight, 0.1 is α4 recent The weights can be adjusted according to actual needs.

[0071] Specifically, if the fourth entropy value of the physiological signal to be identified is less than or equal to the third entry threshold, it indicates good signal quality, and the next step can be performed. If the fourth entropy value of the physiological signal to be identified is greater than the third entry threshold, it indicates poor signal quality, and the next step is not performed. For example, the signal entry standard for the day is the third entry threshold α4. adj_NEW The value is 2.43, if H4 < α4 adj_NEW This indicates good signal quality, which then enables operations such as extracting the heartbeat cycle.

[0072] Wherein, time span n is the short-term time span, indicating a time span of one day, corresponding to the entire period of the day preceding the current date; time span m is the long-term time span, indicating a time span of two days or more, corresponding to the entire period of the two days preceding the current date, or the period prior to two days. Data within time span n represents recent changes in signal characteristics, while data within time span n represents relatively stable changes in signal characteristics after a certain time span of aggregation.

[0073] Among them, the entropy value of the physiological signal within the time period corresponding to the time span n is the entropy value of the physiological signal acquired throughout the previous day.

[0074] Specifically, the third reference threshold satisfies the following formula: α3=H3-H4×K6, where α3 is the third reference threshold, H3 is the third median, H4 is the third standard deviation, and K6 is the sixth adjustment coefficient.

[0075] The calculation formula for the third reference threshold corresponds to the calculation formula (4). The sixth adjustment coefficient is obtained through experiments and can be adjusted appropriately according to the actual situation. In this embodiment, the sixth adjustment coefficient can be 0.4.

[0076] The fourth reference threshold corresponds to α4 in formula (6). adj, the third reference threshold corresponds to α4 in formula (6). recent , a third admission threshold for the fourth entropy value of the physiological signal to be recognized is obtained by performing a weighted sum according to the fourth reference threshold, the weight of the fourth reference threshold, the third reference threshold, and the weight of the third reference threshold. Exemplarily, the weight of the fourth reference threshold is 0.9, and the weight of the third reference threshold is 0.1.

[0077] Among them, both the first time period and the second time period are within the target time period. The first time period is earlier than the second time period. The target time period is the time period corresponding to the time span m. Specifically, the first time period is the whole day of the four days before the current day, and the second time period is the whole day of the three days before the current day. The third threshold is obtained according to formula (6) and is the historical admission threshold actually used. The fourth threshold is obtained according to formula (4), or the fourth threshold is obtained according to formula (4) and formula (5) and is used to adjust the subsequent admission threshold. Finally, the third threshold and the fourth threshold are fused according to formula (6), with 0.9 being the weight of the third threshold and 0.1 being the weight of the fourth threshold.

[0078] Preferably, after obtaining the fourth entropy value and the third admission threshold of the physiological signal, according to the experimental results of the embodiment, the fourth entropy value H4 is used to classify the signals into three categories: the best quality, the second-best quality, and the poor quality. The threshold α4 for the best quality adj_NEW = 2.4, and the threshold for the second-best quality is α4 adj_NEW + 1 = 3.4. The classification rule is defined as follows: The first category with the best quality: classic signal, 0 < H4 < 2.4; the second category with the second-best quality: signal with slight noise, 2.4 < H4 < 3.4; the third category with poor quality: signal with poor quality, 3.4 < H4.

[0079] Among them, please refer to Figures 3-5 , Figure 3 is a schematic diagram of a classic signal provided by an embodiment of the present application, Figure 4 is a schematic diagram of a signal with slight noise provided by an embodiment of the present application, Figure 5 is a schematic diagram of a signal with poor quality provided by an embodiment of the present application. As Figure 3-5 shown, they are all 30-second photoplethysmogram signals with a sampling frequency of 100 Hz. The abscissa is time in seconds, and the ordinate is the AC reading in millivolts. Figure 3 The fourth entropy value of the photoplethysmogram signal shown is 2.1509. According to the above classification rule, it belongs to the first category of physiological signals with the best quality. Figure 4 The fourth entropy value of the photoplethysmogram signal shown is 2.9911. According to the above classification rule, it belongs to the second category of physiological signals with the second-best quality. Figure 5The fourth entropy value of the photoplethysmography signal shown is 3.7152. According to the above classification rule, it belongs to the third category of poor-quality physiological signals.

[0080] Furthermore, the first category can be identified as signal quality meeting the preset standard, or both the first and second categories can be identified as signal quality meeting the preset standard, and then subsequent operations can be performed. The third category of signal quality does not meet the preset standard and can be deleted without performing subsequent operations.

[0081] Furthermore, the third admission threshold can be adjusted based on the hardware performance.

[0082] Please refer to Figure 6 , Figure 6 This is a histogram of the fourth entropy value provided in an embodiment of this application, such as... Figure 6 As shown in the embodiment, from a day of wearing a smartwatch, 1601 valid 30-second photoplethysmography (PPG) signals were screened out according to the above classification rules. The fourth entropy value was calculated for each 30-second PPG signal, and a statistical histogram of the fourth entropy value was obtained, with the horizontal axis representing the entropy value and the vertical axis representing the frequency.

[0083] As can be seen, in this embodiment, performing subsequent operations only when the physiological signal quality meets the requirements helps to eliminate data with high noise or unstable signals, ensuring the reliability and effectiveness of the analyzed data and reducing misjudgments caused by signal quality issues. Simultaneously, adaptively adjusting the third threshold of the fourth entropy value allows for flexible adaptation to dynamic changes in the signal.

[0084] Specifically, the step of extracting the heartbeat cycle from the physiological signal to be identified to obtain a heartbeat cycle sequence includes: sequentially performing filtering, normalization, differentiation, and squaring on the physiological signal to be identified to obtain a reference physiological signal; determining the mean heart rate over the time span n based on the sampling frequency and the second median, where the second median is the median of the heartbeat cycle sequence within the time period corresponding to the time span n; determining a first reference length of the moving window based on the sampling frequency, the mean heart rate, and a fifth adjustment coefficient; obtaining a first length of the moving window within the first time period and a second length of the moving window within the second time period; fusing the first length and the second length to obtain a second reference length of the moving window; fusing the first reference window length and the second reference length to obtain a target length of the moving window; performing a moving average on the reference physiological signal based on the target length to obtain a target physiological signal; performing peak detection on the target physiological signal to obtain multiple wave group positions; determining the time interval between adjacent wave groups in the multiple wave group positions to obtain the heartbeat cycle sequence, where the time interval between a single adjacent wave group corresponds to a single heartbeat cycle.

[0085] Preferably, the Pantompkins algorithm is used for QRS group detection. First, the physiological signal to be identified is preprocessed. For filtering, a bandpass filter is used to remove high-frequency noise and low-frequency baseline drift. This bandpass filter is formed by combining low-pass and high-pass filters, with a frequency range typically from 0.5Hz to 5Hz. Specifically, a low-pass filter with a cutoff frequency of 5Hz is typically used, and the transfer function of the low-pass filter is as follows: Where fc is the cutoff frequency. High-pass filtering typically uses a high-pass filter with a cutoff frequency of 0.5Hz, and the transfer function of the high-pass filter used is as follows: Where fc is the cutoff frequency. Furthermore, the transfer function of the bandpass filter used is as follows: H BP (f)=H LP (f)×H HP (f).

[0086] Furthermore, after filtering, normalization is performed to normalize the filtered signal to a standard range, such as [0, 1]. The specific formula is as follows:

[0087]

[0088] Furthermore, X'[n] represents the normalized physiological signal to be identified, for example, it can represent the normalized photovolume pulse signal.

[0089] Furthermore, the normalized signal is differentiated to enhance the slope of the wavegroup. The specific formula is as follows:

[0090] Y[n] = X′[n+1] - X′[n-1],

[0091] Furthermore, the differential signal is squared to further enhance the amplitude of the wave group. The specific formula is as follows: Z[n] = Y[n] 2 .

[0092] Furthermore, a moving average is applied to the squared signal to smooth it and reduce noise. The specific formula is as follows:

[0093]

[0094] Where L is the length of the moving window, which is exemplified as approximately 100 milliseconds.

[0095] Furthermore, L can be adaptively adjusted using a certain proportional coefficient of the cardiac cycle. Specifically, the average heart rate throughout the day is first calculated using the formula (7): Where Sprate is the sampling frequency, median function represents the median value of the heartbeat cycle sequence, {RR[j]} is the heartbeat cycle sequence for the entire day of day k, and HR day_k Let L be the average heart rate on day k, and then calculate the optimized sliding window length L for that day. day_k The specific calculation formula is (8): Where 0.1 is an adjustment coefficient, obtained from experiments, which can be modified and adjusted according to actual conditions. For example, L day_K = 10 data points (100 milliseconds). Further, refer to calculation formula (9): L adj =mean(L day_k ), L adj The 7-day dynamically adaptive moving window is used as the moving window length for the eighth day. The moving window lengths after the eighth day are weighted and summed, specifically calculated using formula (10): L adj_NEW =L adj ×0.9+L day_k ×0.1, L adj_NEW L represents the adaptive moving window length for the day, indicating the new dynamically adaptive moving window length for the day. adj L is the historical adaptive moving window length calculated two days ago according to formula (10), and L is the actual historical moving window length used. day_k The historical adaptive moving window length, determined one day prior according to formulas (7) and (8), is used as a reference to adjust subsequent moving window lengths. Furthermore, L... day_k It can also be the historical adaptive admission threshold determined one day ago based on calculation formulas (7), (8), and (9); 0.9 is L adj The weight, 0.1 is L day_k The weights can be adjusted according to actual needs.

[0096] Specifically, the first reference length of the moving window satisfies the following formula: L = (sprate / HR) * K5, where K5 is the fifth adjustment coefficient, sprate is the sampling frequency, and HR is the mean heart rate.

[0097] The calculation formula for the first reference length corresponds to calculation formula (8). The fifth adjustment coefficient is obtained through experiments and can be adjusted appropriately according to the actual situation. In this embodiment, the fifth adjustment coefficient can be 0..1.

[0098] Wherein, the first reference length corresponds to L in the calculation formula (10). adj The second reference length corresponds to L in the calculation formula (10). day_kThe optimized moving window length is obtained by weighting and summing the first reference length, its weight, the second reference length, and its weight. For example, the weight of the second reference length is 0.9, and the weight of the first reference length is 0.1.

[0099] The first length is obtained according to calculation formula (10) and is the actual historical admission threshold used. The second length is obtained according to calculation formulas (7) and (8), or the second length is obtained according to calculation formulas (7), (8) and (9) and is used to adjust the subsequent moving window length. Then, referring to calculation formula (10), the second reference length is determined according to the first length, the weight of the first length, the second length and the weight of the second length. Further, the weight of the first length is 0.9 and the weight of the second length is 0.1.

[0100] Furthermore, after the moving average processing, a threshold is set to detect the peak value of the wave group. The initial threshold can be set as a certain percentage of the maximum signal value, such as 50%, i.e., T = 0.5 × min(W[n]). Then, peak values ​​exceeding the threshold are detected in the signal after moving average processing. These peak values ​​correspond to the positions of the wave group. Specifically, if W[n] > T and W[n] > W[n-1] and W[n] > W[n+1], then R[n] = n, where if the signal after moving average processing is greater than the threshold and is greater than the previous and next signals, then the signal is determined to be a peak value, i.e., the position of the wave group.

[0101] Furthermore, the heartbeat cycle is determined based on the aforementioned wave group positions. Specifically, the time interval between adjacent peaks, i.e., the RR interval, is calculated. The formula is as follows: RR[j] = R[j+1] - R[j], where RR[j] represents the time interval of the j-th heartbeat cycle. This yields the time intervals of multiple heartbeat cycles. By sorting the time series, the set of RR intervals is obtained, i.e., the heartbeat cycle sequence.

[0102] As can be seen, in this embodiment, advanced signal processing technology effectively removes noise and interference, accurately identifying the location of each heartbeat and ensuring the reliability of the extracted heartbeat cycle. This allows for accurate calculation of changes in the heartbeat cycle, providing reliable data support for subsequent analysis of irregular cardiac rhythms. Furthermore, during moving average processing, the moving window length is dynamically adjusted by the sampling frequency and average heart rate to smooth the signal, thereby ensuring the reliability of the extracted heartbeat cycle and the difference between adjacent heartbeat cycles.

[0103] S220, determine the first entropy value of the first heartbeat cycle sequence; and determine the first admission threshold of the first entropy value based on a plurality of first reference thresholds determined according to different time spans.

[0104] The first entropy value H1 of the first heartbeat cycle sequence can be determined by referring to the method for determining the fourth entropy value described above. The specific formula is as follows: Where N2 is the total number of RR intervals of the physiological signal to be identified, p RR[j],k2 This represents the probability of RR[n] existing in the k2th interval.

[0105] Specifically, determining the first admission threshold of the first entropy value based on multiple first reference thresholds determined according to different time spans includes: determining the fourth median and fourth standard deviation of multiple sixth entropy values, where a single sixth entropy value is used to indicate the entropy value of a single heartbeat cycle sequence of a single first physiological signal within the time period corresponding to the time span n; determining a fifth reference threshold under the time span n based on the fourth standard deviation, the fourth median, and a seventh adjustment coefficient; obtaining a fifth threshold and a sixth threshold, where the fifth threshold is the threshold for the entropy value of the third heartbeat cycle sequence actually used in the first time period, and the sixth threshold is the threshold for the entropy value of the fourth heartbeat cycle sequence used for reference adjustment in the second time period, where the third heartbeat cycle sequence corresponds to the second physiological signal in the first time period, and the fourth heartbeat cycle sequence corresponds to the third physiological signal in the second time period; fusing the fifth threshold and the sixth threshold to obtain a sixth reference threshold under the time span m; and fusing the fifth reference threshold and the sixth reference threshold to obtain the first admission threshold of the first entropy value.

[0106] Among them, different time spans include time span n and time span m.

[0107] The admission threshold corresponding to the first entropy value is also adaptively adjusted. Specifically, the median of the first entropy value H1 calculated for each heartbeat cycle sequence throughout the day is taken, and the calculation formula is as follows:

[0108] H1 day_k =median(H1) i ),

[0109] The median function represents all heartbeat cycle sequences H1. i After sorting, take the median value, H1 day_k As a representative first entropy value for day k, for example, the first entropy value H1 for day 1. day_1 =1.25.

[0110] Next, the standard deviation of the first entropy value H1 calculated for each heartbeat cycle sequence throughout the day is calculated using the following formula:

[0111] H1 std_k =std(H1) i ),

[0112] Here, the std function represents the computation set {H1}. i Standard deviation, H1 std_k As a standard deviation representing the first entropy value on day k, for example, the standard deviation H1 of the first entropy value on day k. std_1 =0.69.

[0113] Specifically, for the first admission threshold corresponding to the first entropy value, the calculation formula (11) for its adaptive admission threshold is as follows:

[0114] α1 day_k =H1 day_k +H1 std_k ×0.3,

[0115] Where, α1 day_k Let be the admission threshold for day k, where 0.3 is an adjustment coefficient, an empirical threshold that can be adjusted according to actual circumstances. For example, the classification admission threshold α1 for day k. day_1 The value is 1.41. Furthermore, on the eighth day, the average of the first entropy values ​​from the previous seven days, α1, is used as the admission threshold. adj As a signal admission standard, its calculation formula (12) is as follows: α1 adj =mean(α1) day_k ), where α1 day_k The threshold value representing the first entropy value calculated on day k, where k = 1 to 7. The threshold values ​​for the first entropy value after day eight are dynamically weighted, and the calculation formula (13) is as follows:

[0116] α1 adj_NEW =α1 adj ×0.9+α1 recent ×0.1,

[0117] Where, α1 adj_NEW The adaptive admission threshold for the first entropy value of the day, α1 represents the new dynamically adaptive admission threshold for the day. adj The historical adaptive admission threshold calculated two days ago using formula (13) is the actual historical admission threshold used, α1. recent The historical adaptive admission threshold determined one day prior according to formula (11) is not the admission threshold actually used the previous day, and is used to adjust subsequent admission thresholds. Furthermore, α1 recent Alternatively, it can be the historical adaptive admission threshold determined one day prior based on calculation formulas (11) and (12); 0.9 is α1. adj The weight, 0.1 is α1 recent The weights can be adjusted according to actual needs.

[0118] Specifically, the fifth reference threshold satisfies the following formula: α5=H5+H6×K7, where α5 is the fifth reference threshold, H5 is the fourth median, H6 is the fourth standard deviation, and K7 is the seventh adjustment coefficient.

[0119] The calculation formula for the fifth reference threshold corresponds to the calculation formula (11). The seventh adjustment coefficient is obtained through experiments and can be adjusted appropriately according to the actual situation. In this embodiment, the seventh adjustment coefficient can be 0.3.

[0120] The sixth reference threshold corresponds to α1 in formula (13). adj The fifth reference threshold corresponds to α1 in formula (13). recent The first admission threshold for the first entropy value of the first heartbeat cycle sequence is obtained by weighted summation based on the sixth reference threshold, its weight, the fifth reference threshold, and its weight. For example, the weight of the sixth reference threshold is 0.9, and the weight of the fifth reference threshold is 0.1.

[0121] Specifically, the fifth threshold is obtained according to the calculation formula (13), which is the historical admission threshold actually used. The sixth threshold is obtained according to the calculation formula (11), or the sixth threshold is obtained according to the calculation formula (11) and the calculation formula (12), which is used to adjust the subsequent admission threshold. The fifth threshold and the sixth threshold are fused according to the calculation formula (13), with 0.9 as the weight of the fifth threshold and 0.1 as the weight of the sixth threshold.

[0122] As can be seen, in this embodiment of the application, the first admission threshold of the first entropy value is adaptively adjusted, which can flexibly adapt to the dynamic changes of the signal and is conducive to improving the accuracy of monitoring.

[0123] S230, determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain the first cycle difference sequence.

[0124] The difference RRd between adjacent heartbeat cycles is calculated using the following formula: RRd[j] = RR[j+1] - RR[j]. Based on the time series, the differences RRd between all adjacent heartbeat cycles are integrated to obtain the first cycle difference sequence.

[0125] S240, determine the second entropy value of the first period difference sequence; and determine the second admission threshold of the second entropy value based on a plurality of second reference thresholds determined according to different time spans and heart rate variability coefficients.

[0126] The second entropy value H2 of the first period difference sequence can be determined by referring to the method for determining the fourth entropy value described above. The specific formula is as follows: Where N3 is the total number of RRd values ​​of the physiological signal to be identified, p RRd[j],k3 This represents the probability of RR[n] existing in the k3th interval.

[0127] Specifically, the step of determining the second admission threshold of the second entropy value based on multiple second reference thresholds determined according to different time spans and heart rate variability coefficients includes: determining a first standard deviation and a first average value of multiple heartbeat cycle sequences of multiple first physiological signals, wherein the multiple first physiological signals are physiological signals within the time period corresponding to the time span n; determining the heart rate variability coefficient based on the first standard deviation and the first average value; determining a first median and a second standard deviation of multiple third entropy values, wherein a single third entropy value is used to indicate the entropy value of the cycle difference sequence of a single first physiological signal; determining a second reference threshold n under the time span n based on a first adjustment coefficient, the heart rate variability coefficient, the first median, and the second standard deviation; and obtaining a first threshold and a second threshold, wherein the first threshold is the third cycle actually used in the first time period. The threshold for the entropy value of the difference sequence, the second threshold being the threshold for the entropy value of the fourth period difference sequence used for reference adjustment in the second time period, the third period difference sequence corresponding to the second physiological signal in the first time period, the fourth period difference sequence corresponding to the third physiological signal in the second time period, both the first and second time periods being within a target time period, the first time period being earlier than the second time period, the target time period being the time period corresponding to time span m, the time span m being any time span greater than the time span n among the different time spans; the first threshold and the second threshold being fused to obtain a second reference threshold m under the time span m; the second reference threshold n and the second reference threshold m being fused to obtain a second admission threshold for the second entropy value.

[0128] Specifically, the standard deviation and mean of all heart rate cycle sequences for the previous day are calculated, and the heart rate coefficient of variation is obtained based on the ratio of the standard deviation and mean.

[0129] Specifically, determining the heart rate variability coefficient based on the first standard deviation and the first mean includes:

[0130] A first heart rate coefficient of variation is determined based on the first standard deviation and the first average value; a second heart rate coefficient of variation for a third time period and a third heart rate coefficient of variation for a fourth time period are obtained, wherein the third time period is earlier than the time period corresponding to the time span n and the third time period is earlier than the fourth time period; the second heart rate coefficient of variation and the third heart rate coefficient of variation are fused to obtain a fourth heart rate coefficient of variation; the heart rate coefficient of variation is determined based on the first heart rate coefficient of variation and the fourth heart rate coefficient of variation.

[0131] Furthermore, the heart rate variability coefficient of the previous day includes the fourth heart rate variability coefficient based on adaptive adjustment and the first heart rate variability coefficient actually calculated. Weights can be assigned to each of them, and a weighted sum can be performed to calculate the final heart rate variability coefficient of the previous day.

[0132] The process of adaptively adjusting the heart rate coefficient of variation is similar to the process of adaptively adjusting the threshold. The calculation formulas (14) and (15) for the heart rate coefficient of variation are as follows:

[0133] Calculation formula (14):

[0134] Calculation formula (15): HRCV adj =mean(HRCV) day_k ),

[0135] Among them, HRCV day_k represents the average heart rate variability calculated on day k, represents the estimated heart rate variability for that day, {RR[j]} represents the set of RR intervals calculated from all quality-compliant physiological signals throughout the day, i.e., the heart rate cycle sequence, and the mean function represents the average value of the elements within the set. HRCV adj This represents the adaptive heart rate variability coefficient after 7 days of dynamic weighting. For example, the heart rate variability coefficients for days 1-7 are HRCV. day_1 =0.08, HRCV day_2 =0.09, HRCV day_3 =0.07, HRCV day_4 =0.10, HRCV day_5 =0.09, HRCV day_ 6 = 0.07, HRCV day_7 =0.08, weighted heart rate coefficient of variation (HRCV) on day 8 adj =0.083.

[0136] Among them, the dynamically weighted adaptive heart rate variability coefficient (HRCV) after the eighth day adj_NEW The calculation formula (16) is as follows: HRCV adj_NEW =HRCV adj ×0.9+HRCV day_k ×0.1, where HRCV adj_NEW The dynamic adaptive heart rate variability coefficient for the day, HRCV adj The historical adaptive heart rate variability coefficient (HRCV) calculated two days prior using formula (16) is given. day_k The historical adaptive heart rate variability coefficient, determined one day prior according to formula (14), is further... day_kIt can also be the historical adaptive heart rate variability coefficient determined one day ago according to calculation formulas (14) and (15); 0.9 is the HRCV. adj The weight, 0.1 for HRCV day_k The weights can be adjusted according to actual needs.

[0137] Among them, the fourth heart rate variability coefficient is the adaptive heart rate variability coefficient of the previous day. The fourth heart rate variability coefficient is calculated according to the above-mentioned adaptive heart rate variability coefficient calculation method. The third time period corresponds to the current four days in the past. The second heart rate variability coefficient of the third time period is the historical adaptive heart rate variability coefficient calculated according to the calculation formula (16). The fourth time period corresponds to the current three days in the past. The third heart rate variability coefficient of the fourth time period is the historical adaptive heart rate variability coefficient determined according to the calculation formula (14), or it is the historical adaptive heart rate variability coefficient determined according to the calculation formula (14) and the calculation formula (15).

[0138] As can be seen, in this embodiment, adaptive adjustment of the heart rate variability coefficient can more accurately capture newer signal feature changes closely related to physiological state, thereby keenly capturing short-term fluctuations and trend changes in the signal. Simultaneously, the final heart rate variability coefficient is determined based on the adaptively adjusted coefficient and the actually calculated coefficient, comprehensively considering historical data and recent changes, thus adapting more flexibly to dynamic signal changes.

[0139] Specifically, the admission threshold corresponding to the second entropy value is also adaptively adjusted. Specifically, the median of the second entropy value H2 calculated for each period difference sequence throughout the day is taken, and the calculation formula is as follows:

[0140] H2 day_k =median(H2) i ),

[0141] Wherein, the median function represents all periodic difference sequences H2 i After sorting, take the median value, H2 day_k As a representative second entropy value for day k, for example, the second entropy value H2 for day one. day_1 =1.95.

[0142] Next, the standard deviation of the second entropy value H2 calculated for each period difference sequence throughout the day is calculated using the following formula:

[0143] H2 std_k =std(H2) i ),

[0144] Here, the std function represents the computation set {H2}. i Standard deviation, H2 std_kAs a standard deviation representing the second entropy value on day k, for example, the standard deviation H2 of the second entropy value on day 1. std_1 =0.79.

[0145] Specifically, for the first admission threshold corresponding to the first entropy value, the calculation formula (17) for its adaptive admission threshold is as follows:

[0146] α2 day_k =H2 day_k -H2 std_k ×HRCV day_k ×7,

[0147] Where, α2 day_k Let be the admission threshold for day k, where 7 is an adjustment coefficient, an empirical threshold that can be adjusted according to actual circumstances. Furthermore, for day eight, the admission threshold α1 is the average of the second entropy values ​​from the previous seven days. adj As a signal admission standard, its calculation formula (18) is as follows: α2 adj =mean(α2) day_k ), where α2 day_k The threshold value representing the second entropy value calculated on day k, where k = 1 to 7. The threshold values ​​for the second entropy values ​​after day eight are dynamically weighted, and the calculation formula (19) is as follows:

[0148] α2 adj_NEW =α2 adj ×0.9+α2 recent ×0.1,

[0149] Where, α2 adj_NEW The adaptive admission threshold for the second entropy value of the day, α2, represents the new dynamically adaptive admission threshold for the day. adj The historical adaptive admission threshold calculated two days ago using formula (19) is the historical admission threshold actually used two days ago, α2 recent The historical adaptive admission threshold determined one day ago according to formula (17) is not the historical admission threshold actually used one day ago, and is used to adjust subsequent admission thresholds. Furthermore, α2 recent It can also be the historical adaptive admission threshold determined one day ago based on calculation formulas (17) and (18); 0.9 is α2. adj The weight, 0.1 is α2. recent The weights can be adjusted according to actual needs.

[0150] Specifically, the second reference threshold n satisfies the following formula: α2=H1-H2×HRCV×K1, where α2 is the second reference threshold n, K1 is the first adjustment coefficient, H1 is the first median, H2 is the second standard deviation, and HRCV is the heart rate variability coefficient.

[0151] Specifically, the entropy values ​​of all period difference sequences for the entire previous day are calculated, and then the median and standard deviation of the entropy values ​​of all period difference sequences for the entire previous day are determined. Combined with the heart rate coefficient of variation for the previous day, the reference threshold for the previous day, i.e., the second reference threshold n, is determined.

[0152] The calculation formula for the second reference threshold n corresponds to the calculation formula (17). The first adjustment coefficient is obtained through experiments and can be adjusted appropriately according to the actual situation. In this embodiment, the first adjustment coefficient can be 7.

[0153] Wherein, the second reference threshold m corresponds to α2 in the calculation formula (19). adj The second reference threshold n corresponds to α2 in formula (19). recent The second admission threshold for the second entropy value of the periodic difference sequence is obtained by weighted summation of the second reference threshold m, the weight of the second reference threshold m, the second reference threshold n, and the weight of the second reference threshold n. For example, the weight of the second reference threshold m is 0.9, and the weight of the second reference threshold n is 0.1.

[0154] Specifically, the first threshold is obtained according to the calculation formula (19), which is the historical admission threshold actually used. The second threshold is obtained according to the calculation formula (17), or the second threshold is obtained according to the calculation formula (17) and the calculation formula (18), which is used to adjust the subsequent admission threshold. The second threshold and the first threshold are fused according to the calculation formula (19), with 0.9 as the weight of the first threshold and 0.1 as the weight of the second threshold.

[0155] As can be seen, in this embodiment, the signal admission standard threshold is updated by combining a relatively stable second reference threshold m after a certain time span and a second reference threshold n that can reflect newer signal feature changes. This allows the new threshold to comprehensively consider historical data and recent data changes, and to more flexibly adapt to dynamic signal changes. Simultaneously, when calculating the second reference threshold n that reflects newer signal feature changes, the heart rate variability coefficient is added, which can more accurately capture newer signal feature changes closely related to physiological states, thereby keenly capturing short-term fluctuations and trend changes in the signal. Since the heart rate variability coefficient exhibits different characteristics under different physiological or pathological states, incorporating the heart rate variability coefficient into the calculation of signal feature changes can enrich the signal's feature information, improve the recognition of signals under different states, and thus more accurately classify heart rhythms.

[0156] S250, obtain a cardiac rhythm monitoring result according to the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold.

[0157] Specifically, the obtaining of the cardiac rhythm monitoring result according to the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: determining a third classification rule for cardiac rhythm according to the first admission threshold; and determining a fourth classification rule for cardiac rhythm according to the second admission threshold; determining the cardiac rhythm monitoring result according to the third classification rule, the fourth classification rule, the first entropy value, and the second entropy value.

[0158] Exemplarily, wherein, please refer to Figure 7 and Figure 8 , Figure 7 is a histogram of a first entropy value provided by an embodiment of the present application, Figure 8 is a histogram of a second entropy value provided by an embodiment of the present application. As shown in Figure 7 and Figure 8 , calculate the first entropy value and the second entropy value of the 1601 screened photoplethysmogram signals. The first entropy value is the entropy value of the heartbeat period sequence of the photoplethysmogram signal, and the second entropy value is the entropy value of the period difference sequence of the photoplethysmogram signal, and obtain the histograms of the first entropy value and the second entropy value. The abscissa is the entropy value, and the ordinate is the frequency.

[0159] Preferably, according to the experimental results of the embodiment, calculate the first admission threshold and the second admission threshold as 1.4 and 1.4 on the same day. The thresholds between mild and severe irregular cardiac rhythms correspond to 1.4 times the first admission threshold and 1.8 times the second admission threshold as thresholds, and the obtained values are 2.0 and 2.5. Thus, determine the third classification rule and the fourth classification rule.

[0160] The third classification rule is: Cardiac rhythm 1 - Normal cardiac rhythm: H1 ≤ 1.4; Cardiac rhythm 2 - Mildly irregular cardiac rhythm: 1.4 < H1 ≤ 2; Cardiac rhythm 3 - Severely irregular cardiac rhythm: 2 < H1. H1 is the first entropy value.

[0161] The fourth classification rule is: Cardiac rhythm 1 - Normal cardiac rhythm: H2 ≤ 1.4; Cardiac rhythm 2 - Mildly irregular cardiac rhythm: 1.4 < H2 ≤ 2.5; Cardiac rhythm 3 - Severely irregular cardiac rhythm: 2.5 < H2. H2 is the second entropy value.

[0162] Combining the third and fourth classification rules, the general classification rules for cardiac rhythms are obtained: Cardiac rhythm 1 - Normal cardiac rhythm: H1 ≤ 1.4, and H2 ≤ 1.4; Cardiac rhythm 2 - Mildly irregular cardiac rhythm: 1.4

[0163] Furthermore, based on the calculated first entropy value, second entropy value, and overall classification rules, the cardiac rhythm is classified and judged, and the cardiac rhythm monitoring results are output. Among them, the cardiac rhythm monitoring results include at least one of the following: normal cardiac rhythm, mild irregular cardiac rhythm, and severe irregular cardiac rhythm.

[0164] For example, please refer to Figures 9-11 , Figure 9 This is a schematic diagram of a normal heart rhythm provided in an embodiment of this application. Figure 10 This is a schematic diagram of a mildly irregular heart rhythm provided in an embodiment of this application. Figure 11 This is a schematic diagram of a severe irregular heart rhythm provided in an embodiment of this application, as shown below. Figure 9-11 As shown, all are 30-second photoplethysmography (PPG) signals with a sampling frequency of 100Hz. The horizontal axis represents time in seconds, and the vertical axis represents AC readings in millivolts.

[0165] Figure 9 The first entropy value of the heartbeat cycle sequence of the photoplethysmography signal shown is 0.8767, and the second entropy value of the cycle difference sequence is 0.9306. According to the above general classification rules, both the first and second entropy values ​​are less than 1, which meets the first heart rhythm classification and is a normal heart rhythm.

[0166] Figure 10 The first entropy value of the heartbeat cycle sequence of the photoplethysmography signal shown is 1.8473, and the second entropy value of the cycle difference sequence is 2.4089. According to the above general classification rules, both the first and second entropy values ​​are greater than 1.4, which meets the second heart rhythm classification and is a mild irregular heart rhythm. It can be clearly seen from the comparison of the records of the dynamic electrocardiogram monitor that a paroxysmal premature atrial contraction occurred at about 26 seconds.

[0167] Figure 11 The first entropy value of the heartbeat cycle sequence of the photoplethysmography signal shown is 2.0863, and the second entropy value of the cycle difference sequence is 3.1380. According to the above general classification rules, the first entropy value is greater than 2 and the second entropy value is greater than 2.5, which meets the third heart rhythm classification and is a severe irregular heart rhythm. It can be clearly seen from the comparison of the records of the dynamic electrocardiogram monitor that atrial fibrillation occurred in the first 20 seconds, and consecutive premature atrial contractions occurred from the 23rd to the 26th second. ​

[0168] As can be seen, in this embodiment, different classification rules are determined based on different admission thresholds, and then the existence of irregular heart rhythms is judged and classified based on multimodal data. This can more comprehensively reflect the changing characteristics of heart rhythms and improve the accuracy of monitoring. At the same time, a quantitative method is provided to assess the regularity of heart rhythms, which helps to detect heart disease risks early and promotes personalized health management.

[0169] Specifically, before obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold, the method further includes: determining an autonomic nervous system balance index, wherein the autonomic nervous system balance index is used to characterize the degree of variation between the interval differences of adjacent heartbeat cycles; obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: obtaining the cardiac rhythm monitoring result based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold.

[0170] Among them, the autonomic nervous system balance index mainly reflects the balance between sympathetic and parasympathetic nerve activities and is used to accurately assess the state of autonomic nervous function. It is determined by two standard deviation indicators, including short-term variability indicators and long-term variability indicators.

[0171] Specifically, determining the autonomic nervous system balance index includes: determining a second period difference sequence over a second preset time period, the second preset time period including the first preset time period; determining a first variability index and a second variability index for the second period difference sequence, the first variability index indicating the degree of variation among interval differences where the sequence position interval in the second period difference sequence is less than or equal to a preset distance, and the second variability index indicating the degree of variation among interval differences where the sequence position interval in the second period difference sequence is greater than the preset distance; and obtaining the autonomic nervous system balance index based on the ratio of the second variability index to the first variability index.

[0172] Specifically, the first variability index satisfies the following formula: SD1=sqrt(K2*mean((RR[n+1]-RR[n])^2)), where SD1 is the first variability index, K2 is the second adjustment coefficient, and RR[n+1]-RR[n] is the interval difference in the second period difference sequence; the second variability index satisfies the following formula:

[0173] SD2 = sqrt(K3*var(RR)-K4*mean((RR[n+1]-RR[n])^2)), where SD2 is the second variability index, K3 is the third adjustment coefficient, K4 is the fourth adjustment coefficient, and RR is the second heartbeat cycle sequence of the second preset time period.

[0174] The second preset time period is associated with the first preset time period, and for example, it could be the most recent hour within the first preset time period. The first variability index SD1 represents the standard deviation of the difference between adjacent RR intervals, used to measure short-term variability, primarily reflecting the influence of the parasympathetic nervous system (vagus nerve) on the heart. The second variability index SD2 represents the standard deviation of the long-term variation of RR intervals, used to reflect the average difference between all consecutive RR intervals, including not only short-term changes but also long-term changes, thus better representing overall heart rate variability, influenced by both the sympathetic and parasympathetic nervous systems. The autonomic nervous system balance index is the ratio of the second variability index to the first variability index, i.e., SD2 / SD1. A higher ratio may indicate more long-term variability or lower short-term variability, possibly due to increased sympathetic activity or decreased parasympathetic activity.

[0175] To calculate SD1 and SD2, it is first necessary to calculate the period difference sequence, that is, the difference between adjacent RR intervals. Then, according to the calculation formula of SD1 and SD2, SD1 and SD2 are obtained, and then the SD2 / SD1 ratio is determined to obtain the autonomic nervous system balance index.

[0176] Furthermore, based on experimental results, the second adjustment coefficient K2 is 0.5, the third adjustment coefficient K3 is 2, and the fourth adjustment coefficient K4 is 0.5. These adjustment coefficients can be modified to suit specific circumstances.

[0177] Specifically, obtaining the cardiac rhythm monitoring result based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: fusing the first entropy value, the second entropy value, and the autonomic nervous system balance index to obtain the current cardiac state; determining a first classification rule based on a first classification weight, the first admission threshold, and the second admission threshold; determining a second classification rule based on a second classification weight, the first admission threshold, and the second admission threshold; and determining the cardiac rhythm monitoring result based on the current cardiac state, the first classification rule, and the second classification rule.

[0178] Specifically, the current user's heart state P is determined based on the first entropy value, the second entropy value, and the SD2 / SD1 ratio. Furthermore, the current user's heart state P is obtained by weighted summation of the first entropy value, the second entropy value, and the SD2 / SD1 ratio. For example, the weight of the first entropy value is 0.4, the weight of the second entropy value is 0.5, and the weight of the SD2 / SD1 ratio is 0.1.

[0179] The first classification weight is used to distinguish between normal heart rhythms and mildly irregular heart rhythms, while the second classification weight is used to distinguish between mildly irregular heart rhythms and severely irregular heart rhythms. For example, the first classification weight is: a weight of 0.4 for the first admission threshold, a weight of 0.5 for the second admission threshold, and an adjustment coefficient of 0.28. The weighted summation based on these weights and adjustment coefficients yields α. Weighting_Factor1 , representing the first classification rule. The weights for the second classification are: the weight of the first admission threshold is 0.56, the weight of the second admission threshold is 0.9, and the adjustment coefficient is 0.38. The weighted sum of these weights and adjustment coefficients yields α. Weighting_Factor2 , representing the second classification rule. Furthermore, the specific formulas are as follows: α Weighting_Factor1 =α1*0.4+α2*0.5+0.28; α Weighting_Factor2 =α1*0.56+α2*0.9+0.38, where α1 is the first admission threshold and α2 is the second admission threshold.

[0180] The overall classification rule, derived by combining the first and second classification rules, is as follows: A normal heart rhythm is P≦α. Weighting_Factor1 Mild irregular heart rhythm is α Weighting_Factor1 <P≦α Weighting_Factor2 Severe irregular heart rhythm is α Weighting_Factor2 <P。

[0181] Among them, the heart rhythm monitoring results are output according to the general classification rules and the current heart status.

[0182] As can be seen, in this embodiment, the SD2 / SD1 ratio is incorporated when determining and classifying the presence of irregular heart rhythms. Since different types of arrhythmias have different autonomic nervous system regulatory characteristics, this helps to distinguish between different degrees of arrhythmia and improves classification accuracy. Quantifying cardiac status using multimodal data can more comprehensively reflect the changing characteristics of heart rhythms, thereby improving monitoring accuracy. Simultaneously, by calculating and analyzing entropy values ​​in real time, the system can quickly identify irregular heart rhythms and issue warnings, facilitating early intervention and treatment.

[0183] S260, control the display of the heart rhythm monitoring results on the display screen.

[0184] The results of heart rhythm monitoring can be displayed on the screen of a smart wearable device. Please refer to [link / reference]. Figure 12 , Figure 12 This is a schematic diagram of a display screen provided in an embodiment of this application, such as... Figure 12 As shown, the display screen of the smart wearable device shows the results of heart rhythm monitoring. For example, it can show a severely irregular heart rhythm, along with the time and duration of the abnormality, such as atrial fibrillation occurring at time xx, and consecutive premature atrial contractions occurring between the 23rd and 26th seconds of that time.

[0185] Among them, a detailed heart health report is generated based on real-time heart rhythm monitoring results and provided to the user.

[0186] Preferably, smart wearable devices support personalized health management, allowing users to adjust monitoring parameters and warning thresholds based on their specific circumstances.

[0187] Furthermore, smart wearable devices also include communication interfaces for transmitting heart rhythm monitoring results to remote devices, storage media, or doctor terminals, supporting remote diagnosis and consultation. For example, doctors can view a patient's heart rhythm data in real time through a remote monitoring system and adjust treatment plans accordingly. Simultaneously, telemedicine services reduce the number of times patients need to travel to the hospital, saving time and transportation costs.

[0188] Specifically, the method and device can be applied to home health management, enabling self-monitoring in the home environment, helping users understand their heart health status, increasing their sense of security and peace of mind. Through the monitoring results, users can also adjust their lifestyle, such as diet, exercise and rest, thereby improving their overall health.

[0189] This method and device can also be applied in hospitals and clinics for initial screening and long-term follow-up of patients, assisting doctors in diagnosis and treatment. It can also be applied in sports medicine, allowing athletes and fitness enthusiasts to monitor cardiac load during training and prevent exercise-related heart problems. It can enable early detection and warning of irregular heart rhythms, reducing the number of emergency room visits and hospitalizations due to heart disease, and lowering medical costs.

[0190] Furthermore, the method and device can collect a large amount of electrocardiogram signal data for scientific research, providing valuable data support for the development of cardiology. It can also serve as a teaching tool for medical schools and medical institutions, helping students and medical staff to better understand changes in heart rhythm and their clinical significance.

[0191] As can be seen, the embodiments of this application achieve adaptive, high-precision, real-time, and non-invasive irregular heart rhythm detection, which has significant technical advantages and broad application prospects. It not only improves the accuracy of detection and user experience but also provides strong support for the early diagnosis and prevention of cardiovascular diseases, and is of great significance for improving public health.

[0192] For examples consistent with the above embodiments, please refer to... Figure 13 , Figure 13 This is a functional unit block diagram of a heart rhythm monitoring device provided in an embodiment of this application, as shown below. Figure 13 As shown, the heart rhythm monitoring device 130 includes: an extraction unit 131, used to receive a physiological signal to be identified during a first preset time period collected by the optical sensor, and extract the heartbeat cycle from the physiological signal to be identified to obtain a first heartbeat cycle sequence; a first determination unit 132, used to determine a first entropy value of the first heartbeat cycle sequence; and to determine a first admission threshold for the first entropy value based on multiple first reference thresholds determined according to different time spans; a second determination unit 133, used to determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain a first cycle difference sequence; and a third... The determining unit 134 is used to determine the second entropy value of the first period difference sequence; and to determine a second admission threshold for the second entropy value based on a plurality of second reference thresholds determined according to different time spans and heart rate variability coefficients, wherein the heart rate variability coefficient is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans; the monitoring unit 135 is used to obtain the heart rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold and the second admission threshold; and the display unit 136 is used to control the heart rhythm monitoring result to be displayed on the display screen.

[0193] In one possible embodiment, regarding the determination of a second admission threshold for the second entropy value based on multiple second reference thresholds determined according to different time spans and heart rate variability coefficients, the second determining unit 133 is specifically configured to: determine a first standard deviation and a first average of multiple heartbeat cycle sequences of multiple first physiological signals, wherein the multiple first physiological signals are physiological signals within a time period corresponding to the time span n; determine the heart rate variability coefficient based on the first standard deviation and the first average; determine a first median and a second standard deviation of multiple third entropy values, wherein a single third entropy value is used to indicate the entropy value of a cycle difference sequence of a single first physiological signal; determine a second reference threshold n under the time span n based on a first adjustment coefficient, the heart rate variability coefficient, the first median, and the second standard deviation; and obtain a first threshold and a second threshold, wherein the first threshold is the first time period... The threshold for the entropy value of the third period difference sequence actually used, the second threshold is the threshold for the entropy value of the fourth period difference sequence used for reference adjustment in the second time period, the third period difference sequence corresponds to the second physiological signal in the first time period, the fourth period difference sequence corresponds to the third physiological signal in the second time period, both the first time period and the second time period are within the target time period, the first time period is earlier than the second time period, the target time period is the time period corresponding to the time span m, the time span m is any time span among the different time spans that is greater than the time span n; the first threshold and the second threshold are fused to obtain the second reference threshold m under the time span m; the second reference threshold n and the second reference threshold m are fused to obtain the second admission threshold of the second entropy value.

[0194] In one possible embodiment, the second reference threshold n satisfies the following formula: α2=H1-H2×HRCV×K1, where α2 is the second reference threshold n, K1 is the first adjustment coefficient, H1 is the first median, H2 is the second standard deviation, and HRCV is the heart rate variability coefficient.

[0195] In one possible embodiment, before obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold, the cardiac rhythm monitoring device 130 is further configured to: determine an autonomic nervous system balance index, the autonomic nervous system balance index being used to characterize the degree of variation between the interval differences of adjacent heartbeat cycles; obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: obtaining the cardiac rhythm monitoring result based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold.

[0196] In one possible embodiment, in determining the autonomic nervous system balance index, the heart rhythm monitoring device 130 is further configured to: determine a second period difference sequence for a second preset time period, the second preset time period including the first preset time period; determine a first variability index and a second variability index for the second period difference sequence, the first variability index indicating the degree of variation among interval differences where the sequence position interval in the second period difference sequence is less than or equal to a preset distance, and the second variability index indicating the degree of variation among interval differences where the sequence position interval in the second period difference sequence is greater than the preset distance; and obtain the autonomic nervous system balance index based on the ratio of the second variability index to the first variability index.

[0197] In one possible embodiment, the first variability index satisfies the following formula:

[0198] SD1 = sqrt(K2 * mean((RR[n+1] - RR[n])^2)), where SD1 is the first variability index, K2 is the second adjustment coefficient, and RR[n+1] - RR[n] is the interval difference in the second period difference sequence; the second variability index satisfies the following formula: SD2 = sqrt(K3 * var(RR) - K4 * mean((RR[n+1] - RR[n])^2)), where SD2 is the second variability index, K3 is the third adjustment coefficient, K4 is the fourth adjustment coefficient, and RR is the second heartbeat period sequence of the second preset time period.

[0199] In one possible embodiment, in obtaining the heart rhythm monitoring result based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold, the heart rhythm monitoring device 130 is further configured to: fuse the first entropy value, the second entropy value, and the autonomic nervous system balance index to obtain the current heart state; determine a first classification rule based on a first classification weight, the first admission threshold, and the second admission threshold; and determine a second classification rule based on a second classification weight, the first admission threshold, and the second admission threshold; and determine the heart rhythm monitoring result based on the current heart state, the first classification rule, and the second classification rule.

[0200] In one possible embodiment, in obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold, the monitoring unit 135 is further configured to: determine a third classification rule for the cardiac rhythm based on the first admission threshold; and determine a fourth classification rule for the cardiac rhythm based on the second admission threshold; and determine the cardiac rhythm monitoring result based on the third classification rule, the fourth classification rule, the first entropy value, and the second entropy value.

[0201] In one possible embodiment, in extracting the heartbeat cycle from the physiological signal to be identified to obtain a heartbeat cycle sequence, the extraction unit 131 is further configured to: sequentially perform filtering, normalization, differentiation, and squaring on the physiological signal to be identified to obtain a reference physiological signal; determine the mean heart rate over the time span n based on the sampling frequency and the second median, where the second median is the median of the heartbeat cycle sequence within the time period corresponding to the time span n; determine a first reference length of the moving window based on the sampling frequency, the mean heart rate, and a fifth adjustment coefficient; obtain the first length of the moving window within the first time period and the second length of the moving window within the second time period; fuse the first length and the second length to obtain the second reference length of the moving window; fuse the first reference window length and the second reference length to obtain the target length of the moving window; perform moving average processing on the reference physiological signal based on the target length to obtain a target physiological signal; perform peak detection on the target physiological signal to obtain multiple wave group positions; determine the time interval between adjacent wave groups in the multiple wave group positions to obtain the heartbeat cycle sequence, where the time interval between a single adjacent wave group corresponds to a single heartbeat cycle.

[0202] In one possible embodiment, the first reference length of the moving window satisfies the following formula: L = (sprate / HR) * K5, where K5 is the fifth adjustment coefficient, sprate is the sampling frequency, and HR is the mean heart rate.

[0203] In one possible embodiment, before extracting the heartbeat cycle from the physiological signal to be identified to obtain a heartbeat cycle sequence, after receiving the physiological signal to be identified from the optical sensor for a first preset time period, the heart rhythm monitoring device 130 is further configured to: determine the fourth entropy value of the physiological signal to be identified; determine the third median and third standard deviation of multiple fifth entropy values, wherein a single fifth entropy value is used to indicate the entropy value of a single first physiological signal within the time period corresponding to the time span n; determine a third reference threshold under the time span n based on a sixth adjustment coefficient, the third median, and the third standard deviation; obtain a third threshold and a fourth threshold, wherein the third threshold is a threshold for the entropy value of the second physiological signal actually used in the first time period, and the fourth threshold is a threshold for the entropy value of the third physiological signal used for reference adjustment in the second time period; fuse the third threshold and the fourth threshold to obtain a fourth reference threshold under the time span m; fuse the third reference threshold and the fourth reference threshold to obtain a third admission threshold for the fourth entropy value; and detect that the fourth entropy value is less than the third admission threshold.

[0204] In one possible embodiment, the third reference threshold satisfies the following formula: α3 = H3 - H4 × K6, where α3 is the third reference threshold, H3 is the third median, H4 is the third standard deviation, and K6 is the sixth adjustment coefficient.

[0205] In one possible embodiment, in determining the first admission threshold of the first entropy value based on multiple first reference thresholds determined according to different time spans, the first determining unit 132 is further configured to: determine the fourth median and fourth standard deviation of multiple sixth entropy values, wherein a single sixth entropy value is used to indicate the entropy value of a single heartbeat cycle sequence of a single first physiological signal within the time period corresponding to the time span n; determine a fifth reference threshold under the time span n based on the fourth standard deviation, the fourth median, and a seventh adjustment coefficient; obtain a fifth threshold and a sixth threshold, wherein the fifth threshold is a threshold for the entropy value of the third heartbeat cycle sequence actually used in the first time period, and the sixth threshold is a threshold for the entropy value of the fourth heartbeat cycle sequence used for reference adjustment in the second time period, wherein the third heartbeat cycle sequence corresponds to the second physiological signal in the first time period, and the fourth heartbeat cycle sequence corresponds to the third physiological signal in the second time period; fuse the fifth threshold and the sixth threshold to obtain a sixth reference threshold under the time span m; and fuse the fifth reference threshold and the sixth reference threshold to obtain the first admission threshold of the first entropy value.

[0206] In one possible embodiment, the fifth reference threshold satisfies the following formula: α5 = H5 + H6 × K7, where α5 is the fifth reference threshold, H5 is the fourth median, H6 is the fourth standard deviation, and K7 is the seventh adjustment coefficient.

[0207] In one possible embodiment, in determining the heart rate coefficient of variation based on the first standard deviation and the first average, the second determining unit 133 is specifically configured to: determine a first heart rate coefficient of variation based on the first standard deviation and the first average; obtain a second heart rate coefficient of variation for a third time period and a third heart rate coefficient of variation for a fourth time period, wherein the third time period is earlier than the time period corresponding to the time span n, and the third time period is earlier than the fourth time period; fuse the second heart rate coefficient of variation and the third heart rate coefficient of variation to obtain a fourth heart rate coefficient of variation; and determine the heart rate coefficient of variation based on the first heart rate coefficient of variation and the fourth heart rate coefficient of variation.

[0208] It is understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the method embodiment section in this application should be adapted to the device embodiment section in a synchronous manner, and will not be repeated here.

[0209] In the case of using integrated units, please refer to Figure 14 , Figure 14 This is a functional unit block diagram of another cardiac rhythm monitoring device provided in this application embodiment, such as... Figure 14 As shown, the heart rhythm monitoring device 130 includes a processing module 1302 and a communication module 1301. The processing module 1302 controls and manages the operation of the heart rhythm monitoring device 130, for example, executing the steps of the extraction unit 131, the first determination unit 132, the second determination unit 133, the third determination unit 134, the monitoring unit 135, and the display unit 136, and / or performing other processes of the technology described herein. The communication module 1301 is used for interaction between the heart rhythm monitoring device 130 and other devices. Figure 14 As shown, the heart rhythm monitoring device 130 may also include a storage module 1303, which is used to store the program code and data of the heart rhythm monitoring device 130.

[0210] The processing module 1302 can be a processor or controller, such as a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication module 1301 can be a transceiver, RF circuitry, or a communication interface, etc. The storage module 1303 can be a memory.

[0211] All relevant content in each scenario involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here. The above-mentioned heart rhythm monitoring device 130 can perform the above-mentioned... Figure 2 The method for monitoring heart rhythm is shown.

[0212] Please see Figure 15 , Figure 15 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application, as shown below. Figure 15 As shown, the electronic device 1500 includes a processor 1510, a memory 1520, a communication interface 1530, and one or more programs 1521. The one or more programs 1521 are stored in the memory and configured to be executed by the processor. When the program is executed, it includes some or all of the steps of any of the heart rhythm monitoring methods described in the above method embodiments. The processor, memory, and communication interface are interconnected and complete communication between them.

[0213] The memory can be volatile memory such as Dynamic Random Access Memory (DRAM) or non-volatile memory such as a hard disk drive. The memory stores a set of executable program code, and the processor calls the executable program code stored in the memory to execute some or all of the steps of any heart rhythm monitoring method described in the above embodiments of the heart rhythm monitoring method.

[0214] As can be seen, the electronic device 1500 described in this application embodiment first receives the physiological signal to be identified during a first preset time period collected by the optical sensor, and extracts the heartbeat cycle from the physiological signal to be identified to obtain a first heartbeat cycle sequence; then determines a first entropy value of the first heartbeat cycle sequence; and determines a first admission threshold of the first entropy value based on multiple first reference thresholds determined by different time spans; determines the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain a first cycle difference sequence; then determines a second entropy value of the first cycle difference sequence; and determines a second admission threshold of the second entropy value based on multiple second reference thresholds determined by different time spans and the coefficient of variation of heart rate, wherein the coefficient of variation of heart rate is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans; and obtains a heart rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold; finally, controls the heart rhythm monitoring result to be displayed on the display screen.

[0215] This application monitors heart rhythm using multimodal data, including the first entropy value of the heartbeat cycle sequence and the second entropy value of the cycle difference sequence. This can comprehensively reflect the changing characteristics of heart rhythm, meet the high-precision requirements of monitoring, and improve the accuracy of monitoring results. At the same time, when assessing the regularity of heart rhythm, multiple thresholds that are automatically adjusted according to different time spans and / or considering the coefficient of variation of heart rate are added, which is conducive to flexibly adapting to the dynamic changes of physiological signals, thereby further improving the accuracy of monitoring results.

[0216] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0217] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0218] It should be noted that, for the sake of simplicity, the aforementioned methods are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are optional, and the actions and modules involved are not necessarily essential to this application.

[0219] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0220] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0221] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0222] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software program module.

[0223] If the integrated unit is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0224] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk, etc.

[0225] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The above description of the embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for monitoring heart rhythm, characterized in that, A processor for use in a smart wearable device, the smart wearable device further including an optical sensor and a display screen, the method comprising: The system receives the physiological signal to be identified during a first preset time period collected by the optical sensor, and extracts the heartbeat cycle from the physiological signal to be identified to obtain the first heartbeat cycle sequence. Determine a first entropy value for the first heartbeat cycle sequence; and determine a first admission threshold for the first entropy value based on a plurality of first reference thresholds determined for different time spans; Determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain the first cycle difference sequence; Determine a second entropy value for the first period difference sequence; and determine a second admission threshold for the second entropy value based on multiple second reference thresholds determined by different time spans and heart rate variability coefficients, wherein the heart rate variability coefficient is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans; The cardiac rhythm monitoring results are obtained based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold; The heart rhythm monitoring results are displayed on the display screen.

2. The method according to claim 1, characterized in that, The second admission threshold for determining the second entropy value based on multiple second reference thresholds determined according to different time spans and heart rate variability coefficients includes: Determine the first standard deviation and first average of multiple heartbeat cycle sequences of multiple first physiological signals, wherein the multiple first physiological signals are physiological signals within the time period corresponding to the time span n; The heart rate coefficient of variation is determined based on the first standard deviation and the first average value; Determine the first median and second standard deviation of multiple third entropy values, with a single third entropy value used to indicate the entropy value of the period difference sequence of a single first physiological signal; The second reference threshold n for the time span n is determined based on the first adjustment factor, the heart rate coefficient of variation, the first median, and the second standard deviation. A first threshold and a second threshold are obtained. The first threshold is the threshold of the entropy value of the third period difference sequence actually used in the first time period. The second threshold is the threshold of the entropy value of the fourth period difference sequence used for reference adjustment in the second time period. The third period difference sequence corresponds to the second physiological signal in the first time period, and the fourth period difference sequence corresponds to the third physiological signal in the second time period. Both the first time period and the second time period are within the target time period. The first time period is earlier than the second time period. The target time period is the time period corresponding to the time span m. The time span m is any time span among the different time spans that is greater than the time span n. The first threshold and the second threshold are fused to obtain the second reference threshold m for the time span m; The second reference threshold n and the second reference threshold m are fused to obtain the second admission threshold of the second entropy value.

3. The method according to claim 2, characterized in that, The second reference threshold n satisfies the following formula: α2=H1-H2×HRCV×K1, where α2 is the second reference threshold n, K1 is the first adjustment coefficient, H1 is the first median, H2 is the second standard deviation, and HRCV is the heart rate variability coefficient.

4. The method according to claim 1, characterized in that, Before obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold, the method further includes: Determine an autonomic nervous system balance index, which is used to characterize the degree of variability between the intervals of adjacent heartbeat cycles; The step of obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: The cardiac rhythm monitoring results are obtained based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold.

5. The method according to claim 4, characterized in that, The determination of the autonomic nervous system balance index includes: Determine the second periodic difference sequence for a second preset time period, wherein the second preset time period includes the first preset time period; A first variability index and a second variability index are determined for the second period difference sequence. The first variability index is used to indicate the degree of variation among the interval differences in the second period difference sequence where the sequence position interval is less than or equal to a preset distance. The second variability index is used to indicate the degree of variation among the interval differences in the second period difference sequence where the sequence position interval is greater than the preset distance. The autonomic nervous system balance index is obtained based on the ratio of the second variability index to the first variability index.

6. The method according to claim 5, characterized in that, The first variability index satisfies the following formula: SD1=sqrt(K2*mean((RR[n+1]-RR[n])^2)), where SD1 is the first variability index, K2 is the second adjustment coefficient, and RR[n+1]-RR[n] is the interval difference in the second period difference sequence; The second variability index satisfies the following formula: SD2 = sqrt(K3*var(RR)-K4*mean((RR[n+1]-RR[n])^2)), where SD2 is the second variability index, K3 is the third adjustment coefficient, K4 is the fourth adjustment coefficient, and RR is the second heartbeat cycle sequence of the second preset time period.

7. The method according to any one of claims 4-6, characterized in that, The process of obtaining cardiac rhythm monitoring results based on the autonomic nervous system balance index, the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: The current cardiac state is obtained by fusing the first entropy value, the second entropy value, and the autonomic nervous system balance index. A first classification rule is determined based on a first classification weight, a first admission threshold, and a second admission threshold; and a second classification rule is determined based on a second classification weight, a first admission threshold, and a second admission threshold. The cardiac rhythm monitoring result is determined based on the current cardiac state, the first classification rule, and the second classification rule.

8. The method according to any one of claims 1-3, characterized in that, The step of obtaining the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold includes: A third classification rule for determining cardiac rhythm based on the first admission threshold; and a fourth classification rule for determining cardiac rhythm based on the second admission threshold; The cardiac rhythm monitoring result is determined based on the third classification rule, the fourth classification rule, the first entropy value, and the second entropy value.

9. The method according to claim 2, characterized in that, The step of extracting the heartbeat cycle from the physiological signal to be identified to obtain a heartbeat cycle sequence includes: The physiological signal to be identified is sequentially filtered, normalized, differentiated, and squared to obtain a reference physiological signal. The mean heart rate over the time span n is determined based on the sampling frequency and the second median, where the second median is the median of the heart rate cycle sequence within the time period corresponding to the time span n. The first reference length of the moving window is determined based on the sampling frequency, the mean heart rate, and the fifth adjustment coefficient. Obtain the first length of the moving window within the first time period and the second length of the moving window within the second time period; The first length and the second length are merged to obtain the second reference length of the moving window; The target length of the moving window is obtained by fusing the first reference window length and the second reference length. The target physiological signal is obtained by performing a moving average process on the reference physiological signal based on the target length. Peak detection is performed on the target physiological signal to obtain the positions of multiple wave groups; The time interval between adjacent wave groups in the plurality of wave group positions is determined to obtain the heartbeat cycle sequence, and the time interval between a single adjacent wave group corresponds to a single heartbeat cycle.

10. The method according to claim 9, characterized in that, The first reference length of the moving window satisfies the following formula: L=(sprateHR)*K5, where K5 is the fifth adjustment coefficient, sprate is the sampling frequency, and HR is the mean heart rate.

11. The method according to claim 2, characterized in that, Before extracting the heartbeat cycle from the physiological signal to be identified to obtain the heartbeat cycle sequence, after receiving the physiological signal to be identified from the optical sensor during a first preset time period, the method further includes: Determine the fourth entropy value of the physiological signal to be identified; The third median and third standard deviation of multiple fifth entropy values ​​are determined, and a single fifth entropy value is used to indicate the entropy value of a single first physiological signal within the time period corresponding to the time span n; The third reference threshold for the time span n is determined based on the sixth adjustment factor, the third median, and the third standard deviation. Obtain a third threshold and a fourth threshold, wherein the third threshold is the threshold value of the entropy value of the second physiological signal actually used in the first time period, and the fourth threshold is the threshold value of the entropy value of the third physiological signal used for reference adjustment in the second time period; The third threshold and the fourth threshold are fused to obtain the fourth reference threshold under the time span m; The third reference threshold and the fourth reference threshold are fused to obtain the third admission threshold of the fourth entropy value; The fourth entropy value was detected to be less than the third admission threshold.

12. The method according to claim 11, characterized in that, The third reference threshold satisfies the following formula: α3=H3-H4×K6, where α3 is the third reference threshold, H3 is the third median, H4 is the third standard deviation, and K6 is the sixth adjustment coefficient.

13. The method according to claim 2, characterized in that, The first admission threshold for determining the first entropy value based on multiple first reference thresholds determined according to different time spans includes: Determine the fourth median and fourth standard deviation of multiple sixth entropy values, whereby a single sixth entropy value is used to indicate the entropy value of a single heartbeat cycle sequence of a single first physiological signal within the time period corresponding to the time span n; The fifth reference threshold for the time span n is determined based on the fourth standard deviation, the fourth median, and the seventh adjustment factor. A fifth threshold and a sixth threshold are obtained. The fifth threshold is the threshold value of the entropy value of the third heartbeat cycle sequence actually used in the first time period, and the sixth threshold is the threshold value of the entropy value of the fourth heartbeat cycle sequence used for reference adjustment in the second time period. The third heartbeat cycle sequence corresponds to the second physiological signal in the first time period, and the fourth heartbeat cycle sequence corresponds to the third physiological signal in the second time period. The fifth threshold and the sixth threshold are fused to obtain the sixth reference threshold for the time span m; The fifth reference threshold and the sixth reference threshold are fused to obtain the first admission threshold of the first entropy value.

14. The method according to claim 13, characterized in that, The fifth reference threshold satisfies the following formula: α5=H5+H6×K7, where α5 is the fifth reference threshold, H5 is the fourth median, H6 is the fourth standard deviation, and K7 is the seventh adjustment coefficient.

15. The method according to any one of claims 9-14, characterized in that, Determining the heart rate coefficient of variation based on the first standard deviation and the first average includes: The first heart rate variation coefficient is determined based on the first standard deviation and the first average value; Obtain the second heart rate variability coefficient for the third time period and the third heart rate variability coefficient for the fourth time period, wherein the third time period is earlier than the time period corresponding to the time span n, and the third time period is earlier than the fourth time period; The second heart rate coefficient of variation and the third heart rate coefficient of variation are fused to obtain the fourth heart rate coefficient of variation; The heart rate variability coefficient is determined based on the first heart rate variability coefficient and the fourth heart rate variability coefficient.

16. A device for monitoring heart rhythm, characterized in that, A processor for use in a smart wearable device, the smart wearable device also including an optical sensor and a display screen, the device comprising: The extraction unit is used to receive the physiological signal to be identified during a first preset time period collected by the optical sensor, and extract the heartbeat cycle from the physiological signal to be identified to obtain the first heartbeat cycle sequence. The first determining unit is configured to determine a first entropy value of the first heartbeat cycle sequence; and to determine a first admission threshold for the first entropy value based on a plurality of first reference thresholds determined according to different time spans. The second determining unit is used to determine the interval difference between adjacent heartbeat cycles in the first heartbeat cycle sequence to obtain the first cycle difference sequence. The third determining unit is used to determine the second entropy value of the first period difference sequence; and to determine the second admission threshold of the second entropy value based on a plurality of second reference thresholds determined by different time spans and heart rate variability coefficients, wherein the heart rate variability coefficient is related to the time span n, and the time span n is the time span with the smallest time span among the different time spans; The monitoring unit is used to obtain the cardiac rhythm monitoring result based on the first entropy value, the second entropy value, the first admission threshold, and the second admission threshold; The display unit is used to control the display of the heart rhythm monitoring results on the display screen.

17. An electronic device, characterized in that, The device includes: The device includes a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code to perform the steps of the cardiac rhythm monitoring method as described in any one of claims 1-15.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores executable program code, the executable program code including execution instructions for performing the steps of the cardiac rhythm monitoring method as described in any one of claims 1-15.

Images