Object state detection method and device, step-up blood pressure measuring instrument, medium and product

Abstract

The application relates to a state detection method and device of an object, a boost blood pressure measuring instrument, a computer readable storage medium and a computer program product. The method comprises the following steps: determining a pulse interval sequence of the detection object according to pulse data of the detection object in a target time period, performing first feature extraction on the pulse interval sequence to obtain first pulse features, determining a first state of the detection object according to the first pulse features, performing second feature extraction on the pulse interval sequence to obtain second pulse features in the case that the first state is an irregular pulse state, determining a target state of the detection object according to the second pulse features, and displaying on a screen. The method can increase the functionality of the boost blood pressure measuring instrument.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, pressure boosting blood pressure monitor, computer-readable storage medium, and computer program product for detecting the state of an object. Background Technology

[0002] With the continuous development of the medical electronics field and the increasing health awareness of users, more and more portable status monitoring devices are being used by users. Users can use portable status monitoring devices to detect their own status and then make corresponding adjustments based on the detected status.

[0003] Currently, users can obtain their own pulse data through portable status detection devices. However, existing technologies have limited processing methods for pulse data, resulting in relatively limited functionality for portable status detection devices. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, pressure-boosting blood pressure monitor, computer-readable storage medium, and computer program product for detecting the state of an object that can improve the functionality of a pressure-boosting blood pressure monitor, in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a method for detecting the state of an object, comprising: determining a pulse interval sequence of the object based on pulse data of the object within a target time period; extracting a first feature from the pulse interval sequence to obtain a first pulse feature, and determining a first state of the object based on the first pulse feature, wherein the first pulse feature includes time-domain features and nonlinear features; if the first state is an irregular pulse state, extracting a second feature from the pulse interval sequence to obtain a second pulse feature, wherein the second pulse feature includes autocorrelation features and entropy features; determining a target state of the object based on the second pulse feature, and displaying the target state on the screen.

[0006] In one embodiment, a second feature extraction is performed on the pulse interval sequence to obtain a second pulse feature, including: extracting autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence; and extracting entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0007] In one embodiment, autocorrelation features are extracted from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence, including: traversing and extracting the first interval time and the corresponding second interval time in the pulse interval sequence, wherein the lag order of the first interval time and the second interval time is greater than 1; calculating the autocorrelation coefficient of each group of the first interval time and the second interval time; and calculating the autocorrelation features of the pulse interval sequence based on the autocorrelation coefficient of each group.

[0008] In one embodiment, determining the target state of the detected object based on the second pulse feature includes: calculating the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy feature, and the weight corresponding to the entropy feature; and determining the target state based on the target state value.

[0009] In one embodiment, a first pulse feature is obtained by extracting a first feature from the pulse interval sequence, and a first state of the detected object is determined based on the first pulse feature. This includes: extracting multiple time-domain features from the pulse interval sequence and extracting multiple nonlinear features from the pulse interval sequence; determining a time-domain state value corresponding to each time-domain feature based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature; determining a nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature; determining a first state value based on each time-domain state value and each nonlinear state value, and determining a first state based on the first state value.

[0010] In one embodiment, determining the pulse interval sequence of the detection object based on the pulse data of the detection object within a target time period includes: grouping the pulse data according to a preset sliding window to obtain multiple pulse data groups; filtering candidate pulse data in each pulse data group according to preset filtering conditions to obtain multiple candidate pulse peak data; determining the time interval between adjacent candidate pulse peak data; and determining the pulse interval sequence based on each time interval.

[0011] Secondly, this application also provides a state detection device for an object, comprising: a pulse interval sequence determination module, used to determine the pulse interval sequence of the object to be detected based on the pulse data of the object within a target time period; a first feature processing module, used to perform a first feature extraction on the pulse interval sequence to obtain a first pulse feature, and determine a first state of the object to be detected based on the first pulse feature, wherein the first pulse feature includes time-domain features and nonlinear features; a second feature processing module, used to perform a second feature extraction on the pulse interval sequence to obtain a second pulse feature when the first state is an irregular pulse state, wherein the second pulse feature includes autocorrelation features and entropy features; and a state display module, used to determine the target state of the object to be detected based on the second pulse feature, and display the target state on a screen.

[0012] Thirdly, this application also provides a pressure-boosting blood pressure monitor, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in the first aspect above.

[0013] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0014] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect above.

[0015] The aforementioned object's state detection method, apparatus, escalator blood pressure monitor, computer-readable storage medium, and computer program product. While detecting blood pressure in the object, pulse data of the object within a target time period is acquired. Based on the pulse data, the pulse interval sequence of the object is determined. A first feature extraction is performed on the pulse interval sequence to obtain a first pulse feature including time-domain and nonlinear features. Based on the first pulse feature, a first state of the object is determined. This first state detection, followed by the determination of subsequent execution processes, improves computational efficiency. If the first state is an irregular pulse state, a second feature extraction is performed on the pulse interval sequence to obtain a second pulse feature including autocorrelation and entropy features. Based on the second pulse feature, the target state of the object is determined and displayed on the screen. This two-step feature processing enhances the state detection function of the blood pressure monitor, thereby improving the functionality of the escalator blood pressure monitor. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a diagram illustrating the application environment of an object state detection method in one embodiment.

[0018] Figure 2 This is a flowchart illustrating an object state detection method in one embodiment;

[0019] Figure 3 This is a flowchart illustrating step 201 in one embodiment;

[0020] Figure 4 This is a flowchart illustrating step 202 in one embodiment;

[0021] Figure 5 This is a flowchart illustrating step 203 in one embodiment;

[0022] Figure 6 This is a flowchart illustrating step 501 in one embodiment;

[0023] Figure 7 This is a flowchart illustrating step 204 in one embodiment;

[0024] Figure 8 This is a flowchart illustrating the object state detection method in another embodiment;

[0025] Figure 9 This is a structural block diagram of an object state detection device in one embodiment;

[0026] Figure 10 This is a diagram of the internal structure of a booster blood pressure monitor in one embodiment. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0029] The object state detection method provided in this application embodiment can be applied to, for example, Figure 1 The application environment shown includes at least a systolic blood pressure monitor 101.

[0030] The systolic blood pressure monitor 101 is used to collect pulse data of a subject within a target time period and determine the pulse interval sequence. A first feature is extracted from the pulse interval sequence, and a first state is determined based on the extracted first feature. If the first state is an irregular pulse state, a second feature is extracted from the pulse interval sequence, and a target state is determined based on the extracted second feature, which is then displayed on the screen. The systolic blood pressure monitor 101 can be a portable blood pressure monitor powered by a battery.

[0031] The bolus-type blood pressure monitor 101 may include at least one of the following: a cuff, an air pump, a solenoid valve, a pressure sensor, a microcontroller, a signal conditioning circuit, a display screen, a power supply, and an airway. The cuff is connected to the pressure sensor via an air tube, to the solenoid valve via an air tube, and to the air pump via an air tube. The microcontroller may be connected to the pressure sensor, the solenoid valve, the signal conditioning circuit, the display screen, and the power supply. The signal conditioning circuit may be connected to the solenoid valve, the pressure sensor, the microcontroller, the signal conditioning circuit, the display screen, and the power supply.

[0032] In real-world scenarios, atrial fibrillation is a common and dangerous arrhythmia, and early screening helps ensure the user's health. Oscillometric blood pressure monitors capture cuff oscillations during inflation / deflation, containing heart rate and pulse rhythm information, but traditional devices often ignore this information. Existing blood pressure monitors with arrhythmia alerts mostly use single variability indicators, which are prone to misinterpretation by events such as premature beats, and often lack the specificity to achieve high-specificity detection on resource-constrained microcontrollers. To address this, this application utilizes a multi-feature fusion approach for state detection within the conventional measurement process, making it suitable for embedded, lightweight devices and improving the functionality of booster blood pressure monitors.

[0033] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting the state of an object is provided, which can be applied to... Figure 1 The following explanation is based on a vasopressor blood pressure monitor, including steps 201 to 204.

[0034] Step 201: Determine the pulse interval sequence of the subject based on the pulse data of the subject within the target time period.

[0035] In this application, pulse data refers to the periodic pressure fluctuation signal generated by the minute vibrations of the arterial wall caused by the heartbeat during the gradual increase of cuff pressure in a vasopressor, transmitted through the cuff bladder to the pressure sensor; that is, pulse wave oscillation data. Pulse data is periodic, with time intervals between pulse peaks. The pulse data can be analog data, such as continuous waveform data, or scattered data, such as a binary tuple including time and intensity.

[0036] During implementation, the systolic blood pressure monitor collects pulse data from the subject within a target time period. The pulse data is then segmented and filtered. Based on the time intervals between pulse peaks obtained after segmentation and filtering, a pulse interval sequence for the subject is constructed. The target time period can be the time it takes for the systolic blood pressure monitor to perform a single blood pressure measurement, and this time can be preset.

[0037] During execution, the systolic blood pressure monitor can be initialized first. During initialization, the detector and buffer can be initialized. After initialization, the systolic blood pressure monitor collects pulse wave oscillation data of the test subject within a preset time period through the pressure sensor. During the acquisition process, the systolic blood pressure monitor can continuously sample the pressure signal at a fixed sampling rate and perform necessary bandpass / lowpass filtering to remove power frequency noise and DC drift.

[0038] During the construction of the pulse interval sequence, the pulse interval sequence can be constructed based on the time interval between pulse peaks. Since the pulse intensity of each detection object is different, the peak range of the pulse peak can be determined based on the pulse data. Candidate pulse peaks in the pulse data can be determined based on the peak range. The pulse interval sequence can be constructed based on the time interval between the candidate pulse peaks.

[0039] Step 202: Extract the first feature from the pulse interval sequence to obtain the first pulse feature, and determine the first state of the detection object based on the first pulse feature.

[0040] In real-world scenarios, portable blood pressure monitors typically have limited computing power. Therefore, the allocation of computing power is crucial. During the determination of the object's state, the state can be determined in stages. Subsequent calculations are performed only if the object's first state is an irregular pulse state; otherwise, no further calculations are performed. This improves the utilization rate of the blood pressure monitor's computing power. The first state characterizes whether the pulse of the object is regular; the first pulse characteristics include temporal and nonlinear features.

[0041] In this application, time-domain features are used to characterize the characteristics or patterns of pulse data in the time domain, and time-domain features may include multiple types; nonlinear features are used to characterize the heart rate variability of pulse data, and nonlinear features may include multiple types.

[0042] During implementation, the vasopressor blood pressure monitor extracts and processes the first feature of each interval in the pulse interval sequence to obtain time-domain and nonlinear features. Based on these features, feature calculations are performed to determine the first state of the subject. If the first state of the subject is an irregular pulse state, step 203 is executed; otherwise, the first state is determined to be a normal state. The irregular pulse state can be an arrhythmic pulse state or an abnormal pulse rhythm state, used to characterize the lack of regularity in the time interval, intensity, or rhythm of the subject's arterial pulsation.

[0043] Furthermore, when the first state is normal, the systolic blood pressure monitor can display the normal state on the screen, and the screen can remain lit and display the normal state after the measurement operation of the systolic blood pressure monitor is completed.

[0044] Step 203: Extract the second feature from the pulse interval sequence to obtain the second pulse feature.

[0045] During implementation, the vasopressor blood pressure monitor extracts and processes the second feature of each interval in the pulse interval sequence to obtain autocorrelation features and entropy features. The second pulse features include both autocorrelation and entropy features.

[0046] In the process of extracting entropy features, the vasopressor blood pressure monitor can extract sample entropy from each interval time in the pulse interval sequence to obtain sample entropy, and extract Shannon entropy from each interval time in the pulse interval sequence to obtain Shannon entropy; among them, sample entropy is used to characterize the complexity of the pulse interval sequence, and Shannon entropy is used to characterize the uncertainty or information content of the pulse interval sequence.

[0047] Step 204: Determine the target state of the detection object based on the second pulse characteristics, and display the target state on the screen.

[0048] During implementation, the vasopressor blood pressure monitor performs feature calculations on the second pulse characteristic to obtain the target state value of the subject. Based on the target state value, the target state of the subject is determined and displayed on the screen. The target state characterizes the atrial fibrillation state of the subject, and includes either atrial fibrillation or irregular pulse state.

[0049] In addition, the vasopressor blood pressure monitor can determine the display method according to the target state. When the target state is atrial fibrillation, it can flash the atrial fibrillation state on the screen and announce the atrial fibrillation state via voice. When the target state is normal, it can keep the normal state on the screen. When the target state is irregular pulse, it can keep the irregular state on the screen.

[0050] In the aforementioned object state detection method, while performing blood pressure detection on the object, the pulse data of the object within the target time period is first acquired. Based on the pulse data, the pulse interval sequence of the object is determined. The pulse interval sequence is then subjected to first feature extraction to obtain first pulse features including time-domain features and nonlinear features. Based on the first pulse features, the first state of the object is determined. By performing first state detection first, the subsequent execution process is determined, improving the utilization rate of computing power. When the first state is an irregular pulse state, the pulse interval sequence is subjected to second feature extraction to obtain second pulse features including autocorrelation features and entropy features. Based on the second pulse features, the target state of the object is determined and displayed on the screen. Through two feature processing steps, the state detection function of the blood pressure measuring instrument is increased, thereby improving the functionality of the bolus blood pressure measuring instrument.

[0051] Based on the above exemplary embodiment, the following provides a method for detecting the state of an object in one or more exemplary embodiments, which is applied to... Figure 1 Taking the systolic blood pressure monitor as an example, the explanation includes the following:

[0052] In determining the pulse interval sequence, multiple pulse data groups can be determined according to a preset sliding window, and pulse peak data can be filtered from the multiple pulse data groups. Based on the time interval between each pulse peak data, a pulse interval sequence is constructed. In one optional implementation provided by this application, such as... Figure 3 As shown, step 201 includes steps 301 to 303:

[0053] Step 301: Group the pulse data according to the preset sliding window to obtain multiple pulse data groups.

[0054] During implementation, the vasopressor blood pressure monitor groups the pulse data according to the size of the preset sliding window, resulting in multiple pulse data groups.

[0055] Each pulse data set may include multiple scattered pulse data points, or a segment of simulated pulse data. During execution, a sliding window with an odd number of window sizes can be used. The sliding window duration can be 10ms, 5ms, or 15ms.

[0056] Step 302: Filter the candidate pulse data in each pulse data group according to the preset filtering conditions to obtain multiple candidate pulse peak data.

[0057] In real-world scenarios, it is more convenient to calculate and the data is more reliable when the pulse peak is at the center point of the sliding window. To address this, we can first filter each pulse data group to obtain a candidate pulse data group that includes candidate pulse data, and then extract the pulse peaks from the candidate pulse data group to obtain multiple candidate pulse peak data.

[0058] During implementation, the vasopressor blood pressure monitor compares the center of the window with other sample points at each sampling time for each pulse data set. If the center is the maximum value within the window, the pulse data set is designated as a candidate pulse data set. After the candidate pulse data sets are determined, the maximum value, i.e., the pulse peak, is extracted from the candidate pulse data sets, resulting in multiple candidate pulse peak data. One selection criterion is that the center of the sliding window is the maximum value within the window.

[0059] Step 303: Determine the time interval between adjacent candidate pulse peak data, and determine the pulse interval sequence based on each time interval.

[0060] In real-world scenarios, there are also instances where candidate peak selection is flawed. For example, a candidate peak within a sliding window may not be the peak value of an arterial pulsation. This can result in a small time interval between pulse peaks, affecting the determination of the target state. To address this, after determining multiple time intervals, the median corresponding to each time interval can be determined. The time intervals can then be filtered based on the median to obtain the target time interval. Finally, a pulse interval sequence can be constructed based on the target time interval.

[0061] During implementation, the vasopressor blood pressure monitor calculates the time interval between each two adjacent candidate pulse peak data. After obtaining multiple time intervals, it determines the median of each time interval, determines the target interval based on the median, selects the target time interval within the target interval from each time interval, and constructs a pulse interval sequence based on the target time interval.

[0062] During execution, the vasopressor blood pressure monitor can calculate the time difference between adjacent pulse peaks based on candidate pulse peak data to obtain a pulse interval sequence, i.e., the RR sequence. The RR sequence is first filtered using the median and median absolute deviation (MAD). For example, the lower limit of the interval is set as median - coefficient * median absolute deviation, and the upper limit of the interval is set as median + coefficient * median absolute deviation. Further, intervals that exceed the range are removed, and only the accepted intervals are used to calculate the pulse rate or as feature input. If all intervals are removed, the pulse rate is estimated using the median. The coefficient can be 2.

[0063] One optional implementation provided in this application constructs a pulse interval sequence by filtering pulse interval times, thereby avoiding the influence of erroneous data on the pulse interval sequence and improving the accuracy of the first state and the target state.

[0064] In the first pulse feature processing, a time-domain state value can be calculated for each extracted time-domain feature, and a non-linear state value can be calculated for each extracted non-linear feature. Based on each time-domain feature value and each non-linear state value, the first state of the detected object is determined. In one optional implementation provided by this application, such as… Figure 4 As shown, step 202 includes steps 401 to 404:

[0065] Step 401: Extract time-domain features from the pulse interval sequence to obtain multiple time-domain features, and extract nonlinear features from the pulse interval sequence to obtain multiple nonlinear features.

[0066] In real-world scenarios, temporal features are used to characterize the characteristics of each interval in the pulse interval sequence, while nonlinear features are used to characterize the heart rate variability of the detected object. Whether the pulse / heart rate of the detected object is regular can be determined through temporal and nonlinear features.

[0067] During implementation, the vasopressor blood pressure monitor performs multiple rounds of time-domain feature extraction on multiple intervals in the pulse interval sequence to obtain multiple time-domain features, and performs multiple rounds of nonlinear feature extraction on multiple intervals in the pulse interval sequence to obtain multiple nonlinear features.

[0068] During execution, the systolic blood pressure monitor can calculate the mean, standard deviation, root mean square of adjacent intervals, heart rate variability, and coefficient of variation (standard deviation divided by mean) for multiple intervals in the pulse interval sequence, obtaining the mean, standard deviation, root mean square of adjacent intervals, heart rate variability, and coefficient of variation as time-domain features; and the systolic blood pressure monitor can calculate the scatter plot indices of the Poincaré plot for multiple intervals in the pulse interval sequence, namely: the standard deviation of the minor axis of the ellipse (SD1), the standard deviation of the major axis of the ellipse (SD2), and the ratio of the standard deviation of the minor axis of the ellipse to the standard deviation of the major axis of the ellipse (SD1 / SD2) as nonlinear features.

[0069] Step 402: Based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature, determine the time-domain state value corresponding to each time-domain feature.

[0070] During implementation, for each time-domain feature, the vasopressor blood pressure monitor calculates the relationship between the first feature value and the corresponding first threshold value of the time-domain feature, and determines the initial state value and weight corresponding to the relationship. Based on the initial state value and weight, the time-domain state value corresponding to the time-domain feature is determined until the time-domain state value of all time-domain features is calculated.

[0071] During execution, for any given time-domain feature, the vasopressor blood pressure monitor can detect whether the first feature value of the time-domain feature is greater than the corresponding first threshold. If it is greater, an initial state value corresponding to the first threshold is determined; if it is less, the initial state value is determined to be 0. After determining the initial state value, the preset weight of the time-domain feature is obtained, and the initial state value and the weight are multiplied to obtain the time-domain state value. The initial state values ​​of all time-domain features can be the same, in which case the weights of each time-domain feature are different; or, the weights of all time-domain features can be the same, in which case the initial state values ​​of each time-domain feature are different.

[0072] For example, when the first feature value is greater than the corresponding first threshold, the initial state value is determined to be 1 and the corresponding weight is x1%, then the time domain state value of this time domain feature is x1;

[0073] For example, when the first feature value is greater than the corresponding first threshold, the initial state value is determined to be x2 and the corresponding weight is 1, then the time-domain state value of this time-domain feature is x2.

[0074] Step 403: Based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature, determine the nonlinear state value corresponding to each nonlinear feature.

[0075] During implementation, for each nonlinear feature, the systolic blood pressure monitor calculates the relationship between the second feature value and the corresponding second threshold of the nonlinear feature, and determines the initial state value and weight corresponding to the relationship. Based on the initial state value and weight, the nonlinear state value corresponding to the nonlinear feature is determined until the nonlinear state values ​​of all nonlinear features are calculated.

[0076] During execution, for any nonlinear feature, the vasopressor blood pressure monitor can detect whether the first characteristic value of the nonlinear feature is greater than the corresponding first threshold. If it is greater, an initial state value corresponding to the first threshold is determined; if it is less, the initial state value is determined to be 0. After determining the initial state value, the preset weight of the nonlinear feature is obtained, and the initial state value and the weight are multiplied to obtain the time-domain state value. The initial state values ​​of each nonlinear feature can be the same, in which case the weights of each nonlinear feature are different; or, the weights of each nonlinear feature can be the same, in which case the initial state values ​​of each nonlinear feature are different.

[0077] Step 404: Determine the first state value based on each time-domain state value and each nonlinear state value, and determine the first state based on the first state value.

[0078] During implementation, the bolus blood pressure monitor can sum up each time-domain state value and each nonlinear state value to obtain the first state value, and determine the first state based on the first state value.

[0079] During the determination of the first state, the vasopressor blood pressure monitor can detect whether the first state value is greater than the first state value threshold. If the first state value is greater than the first state value threshold, it indicates that the pulse state of the tested object is irregular, and the first state is determined to be an irregular pulse state. If the first state value is less than the first state value threshold, it indicates that the pulse state of the tested object is regular, and the first state is determined to be a normal state.

[0080] One optional implementation provided in this application sets thresholds and weights for each time-domain feature and nonlinear feature, avoiding misjudgment of the state of the detected object, improving the accuracy and reliability of the first state, and thus improving the reliability of the bolus blood pressure monitor in the process of function implementation.

[0081] In the second feature processing, the vasopressor blood pressure monitor can extract autocorrelation features from the pulse interval sequence to obtain autocorrelation features, and extract entropy features from the pulse interval sequence to obtain entropy features; in one optional embodiment provided by this application, such as Figure 5As shown, step 203 includes steps 501 to 502:

[0082] Step 501: Extract autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence.

[0083] During implementation, the vasopressor blood pressure monitor extracts the autocorrelation coefficients of each interval in the pulse interval sequence, obtaining multiple autocorrelation coefficients. Based on these autocorrelation coefficients, feature fusion processing is performed to obtain the autocorrelation features of the pulse interval sequence.

[0084] Step 502: Extract entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0085] During implementation, the vasopressor blood pressure monitor calculates the sample entropy of each interval in the pulse interval sequence to obtain the sample entropy of the pulse interval sequence, and calculates the Shannon entropy of each interval in the pulse interval sequence to obtain the Shannon entropy of the pulse interval sequence.

[0086] An optional implementation provided in this application further refines the process of determining atrial fibrillation state by extracting autocorrelation features and entropy features from each interval time in the pulse interval sequence. This avoids misjudgment caused by time-domain feature determination, improves the reliability of atrial fibrillation state, and thus improves the reliability of the vasopressor blood pressure monitor.

[0087] During the calculation of the autocorrelation coefficient, premature beats may cause interference. Premature beats refer to the occasional, random early contractions of the main pulse, which are different from atrial fibrillation. However, premature beat data can affect the assessment of atrial fibrillation status. To address this, the autocorrelation coefficient of the pulse interval sequence can be calculated based on a larger lag order to avoid the influence of premature beats on the atrial fibrillation status. In one optional implementation provided in this application, such as... Figure 6 As shown, step 501 includes steps 601 to 603:

[0088] Step 601: Iterate through and extract the first interval time and the corresponding second interval time from the pulse interval sequence.

[0089] During implementation, the systolic blood pressure monitor iterates through and extracts the first interval time and the second interval time corresponding to the lag order in the pulse interval sequence. During execution, for each first interval time, the systolic blood pressure monitor determines the second interval time corresponding to the lag order, and extracts all combinations of first and second interval times. The lag order of the first and second interval times is greater than 1; that is, the first and second interval times are not consecutive in the pulse interval sequence. Preferably, the lag order is 2, 3, 4, or 5.

[0090] In this application, the first interval time and the second interval time can be a data set containing multiple consecutive interval times; wherein, the first and second interval times in the first interval time and the first and second interval times are not consecutive in the pulse interval sequence, in order to avoid the influence of premature beats on the atrial fibrillation state, thereby improving the reliability of the target state.

[0091] For example, if the first interval is [t0, t1, t2], then the second interval can be [t2, t3, t4].

[0092] Step 602: Calculate the autocorrelation coefficient for the first and second intervals of each group.

[0093] During implementation, the systolic blood pressure monitor calculates the autocorrelation coefficient of the first and second intervals for each combination of the first and second intervals.

[0094] Step 603: Calculate the autocorrelation characteristics of the pulse interval sequence based on the autocorrelation coefficients of each group.

[0095] During implementation, the vasopressor blood pressure monitor performs feature fusion on the autocorrelation coefficients of each group to obtain the autocorrelation features of the pulse interval sequence, and uses the autocorrelation features to characterize the correlation of each time interval in the pulse interval sequence.

[0096] One optional implementation provided in this application avoids the influence of premature beats on atrial fibrillation by increasing the lag order of autocorrelation, thereby improving the accuracy of the vasopressor blood pressure monitor in determining atrial fibrillation. Furthermore, by calculating the autocorrelation characteristics only by adjusting the lag order, computational power is saved, allowing the vasopressor blood pressure monitor to perform calculations quickly while avoiding the influence of premature beats, thus improving the accuracy of the target state.

[0097] In determining the target state, the vasopressor blood pressure monitor can first determine the target state value of the detected object based on the first pulse characteristic, the weight corresponding to the first pulse characteristic, the second pulse characteristic, and the weight corresponding to the second pulse characteristic, and then determine the target state based on the target state value; in one optional embodiment provided in this application, such as Figure 7 As shown, step 204 includes steps 701 to 702:

[0098] Step 701: Calculate the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy feature, and the weight corresponding to the entropy feature.

[0099] During implementation, the vasopressor blood pressure monitor can query autocorrelation features, the weights corresponding to the autocorrelation features, entropy features, and the weights corresponding to the entropy features. It can detect whether the feature values ​​of the autocorrelation features are greater than the corresponding thresholds, determine the initial autocorrelation feature values ​​based on their magnitude relationships, and detect whether the feature values ​​of the entropy features are greater than the corresponding thresholds, determine the initial entropy feature values ​​based on their magnitude relationships, and perform a weighted summation based on the autocorrelation feature values, the weights corresponding to the autocorrelation features, the entropy feature values, and the weights corresponding to the entropy features to obtain the target state value.

[0100] For any autocorrelation feature or entropy-type feature, the vasopressor blood pressure monitor can detect whether the feature value is greater than the corresponding threshold. If it is greater, the initial state value corresponding to the feature is determined; if it is less, the initial state value corresponding to the feature is determined to be 0. The state value corresponding to the feature is then determined based on the weight of the feature and the initial state value. The initial state values ​​of all features can be the same, in which case the weights of each feature are different; or the weights of all features can be the same, in which case the initial state values ​​of each feature are different. Specific examples can be found in the example of time-domain features described above, and will not be repeated in this embodiment.

[0101] Step 702: Determine the target state based on the target state value.

[0102] During implementation, the vasopressor blood pressure monitor can detect whether the target state value is greater than the state value threshold. If it is greater, it indicates that the subject has atrial fibrillation and the target state is determined to be atrial fibrillation. If it is less than, it indicates that the subject does not have atrial fibrillation, but the pulse is irregular and the target state is determined to be irregular pulse.

[0103] One optional implementation method provided in this application uses a threshold judgment method to simply and accurately determine the target state, avoiding complex calculations, saving computing power, and ensuring the accuracy of the target state.

[0104] In one embodiment, see Figure 8 The diagram illustrates a flowchart of an object state detection method provided in an embodiment of this application. This object state detection method can be applied to... Figure 1 In the example shown, a pressure-boosting blood pressure monitor. For instance... Figure 8 As shown, the state detection method for this object may include the following steps:

[0105] Step 801: Collect pulse data of the target object during the target time period.

[0106] Step 802: Group the pulse data according to the preset sliding window to obtain multiple pulse data groups.

[0107] Step 803: Filter the candidate pulse data in each pulse data group according to the preset filtering conditions to obtain multiple candidate pulse peak data.

[0108] Step 804: Determine the time interval between adjacent candidate pulse peak data, and determine the pulse interval sequence based on each time interval.

[0109] Step 805: Extract time-domain features from the pulse interval sequence to obtain multiple time-domain features, and extract nonlinear features from the pulse interval sequence to obtain multiple nonlinear features.

[0110] Step 806: Determine the time-domain state value corresponding to each time-domain feature based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature.

[0111] Step 807: Determine the nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature.

[0112] Step 808: Determine the first state value based on each time-domain state value and each nonlinear state value, and determine the first state based on the first state value.

[0113] Step 809: In the case of an irregular pulse state in the first state, iterate through and extract the first interval time and the corresponding second interval time in the pulse interval sequence.

[0114] Step 810: Calculate the autocorrelation coefficients of the first and second interval times for each group, and calculate the autocorrelation characteristics of the pulse interval sequence based on the autocorrelation coefficients of each group.

[0115] Step 811: Extract entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0116] Step 812: Calculate the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy feature, and the weight corresponding to the entropy feature, and determine the target state based on the target state value.

[0117] It should be noted that any one or more of steps 801 to 812 can be combined to form a new implementation method according to the needs of implementation and deployment. Furthermore, any one or more technical features in the technical solution composed of steps 801 to 812 can also be combined to form a new implementation method according to the actual deployment needs, or technical features in one or more optional implementations provided by one or more of the above embodiments can be combined to form a new implementation method. These will not be elaborated on here.

[0118] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0119] Based on the same inventive concept, this application also provides an object state detection device for implementing the object state detection method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in the one or more object state detection device embodiments provided below can be found in the limitations of the object state detection method described above, and will not be repeated here.

[0120] In one exemplary embodiment, such as Figure 9As shown, an object state detection device is provided, comprising: a pulse interval sequence determination module 901, a first feature processing module 902, a second feature processing module 903, and a state display module 904, wherein: the pulse interval sequence determination module 901 is used to determine the pulse interval sequence of the detected object based on the pulse data of the detected object within a target time period; the first feature processing module 902 is used to perform a first feature extraction on the pulse interval sequence to obtain a first pulse feature, and determine a first state of the detected object based on the first pulse feature, wherein the first pulse feature includes time-domain features and nonlinear features; the second feature processing module 903 is used to perform a second feature extraction on the pulse interval sequence to obtain a second pulse feature when the first state is an irregular pulse state, wherein the second pulse feature includes autocorrelation features and entropy features; and the state display module 904 is used to determine the target state of the detected object based on the second pulse feature and display the target state on the screen.

[0121] In one embodiment, the second feature processing module 903 includes an autocorrelation feature processing unit and an entropy feature processing unit, wherein: the autocorrelation feature processing unit is used to extract autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence; the entropy feature processing unit is used to extract entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0122] In one embodiment, the autocorrelation feature processing unit includes an interval time extraction unit, an autocorrelation coefficient calculation unit, and an autocorrelation feature determination unit, wherein: the interval time extraction unit is used to traverse and extract the first interval time and the corresponding second interval time in the pulse interval sequence, and the lag order of the first interval time and the second interval time is greater than 1; the autocorrelation coefficient calculation unit is used to calculate the autocorrelation coefficient of each group of first interval time and second interval time; and the autocorrelation feature determination unit is used to calculate the autocorrelation feature of the pulse interval sequence based on the autocorrelation coefficient of each group.

[0123] In one embodiment, the state display module 904 includes a target state value determination unit and a target state determination unit. The target state value determination unit is used to calculate the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy feature, and the weight corresponding to the entropy feature. The target state determination unit is used to determine the target state based on the target state value.

[0124] In one embodiment, the first feature processing module 902 includes a feature extraction unit, a temporal feature processing unit, a nonlinear feature processing unit, and a first state determination unit, wherein: the feature extraction unit is used to extract temporal features from the pulse interval sequence to obtain multiple temporal features, and to extract nonlinear features from the pulse interval sequence to obtain multiple nonlinear features; the temporal feature processing unit is used to determine the temporal state value corresponding to each temporal feature based on the relationship between the first feature value and the corresponding first threshold of each temporal feature, and the weight of each temporal feature; the nonlinear feature processing unit is used to determine the nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature; the first state determination unit is used to determine a first state value based on each temporal state value and each nonlinear state value, and to determine a first state based on the first state value.

[0125] In one embodiment, the pulse interval sequence determination module 901 includes a pulse data group determination unit, a candidate pulse peak determination unit, and a pulse interval sequence construction unit, wherein: the pulse data group determination unit is used to group pulse data according to a preset sliding window to obtain multiple pulse data groups; the candidate pulse peak determination unit is used to filter candidate pulse data in each pulse data group according to preset filtering conditions to obtain multiple candidate pulse peak data; and the pulse interval sequence construction unit is used to determine the time interval between adjacent candidate pulse peak data and determine the pulse interval sequence based on each time interval.

[0126] Each module in the aforementioned status detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the bolus blood pressure monitor in hardware form or independent of it, or stored in the memory of the bolus blood pressure monitor in software form, so that the processor can call and execute the corresponding operations of each module.

[0127] In one exemplary embodiment, a booster blood pressure monitor is provided. This booster blood pressure monitor can be a terminal, and its internal structure diagram can be as follows: Figure 10As shown. The systolic blood pressure monitor may include at least one of a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of the systolic blood pressure monitor provides computational and control capabilities. The memory of the systolic blood pressure monitor includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the systolic blood pressure monitor is used for exchanging information between the processor and external devices. The communication interface of the systolic blood pressure monitor is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for detecting the state of an object. The display unit of this bolus blood pressure monitor is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of this bolus blood pressure monitor can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad set on the casing of the bolus blood pressure monitor, or an external keyboard, touchpad, or mouse, etc.

[0128] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the systolic blood pressure monitor to which the present application is applied. A specific systolic blood pressure monitor may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0129] In an exemplary embodiment, a vasopressor blood pressure monitor is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: determining a pulse interval sequence of the target object based on pulse data of the target object within a target time period; performing a first feature extraction on the pulse interval sequence to obtain a first pulse feature, and determining a first state of the target object based on the first pulse feature, wherein the first pulse feature includes time-domain features and nonlinear features; if the first state is an irregular pulse state, performing a second feature extraction on the pulse interval sequence to obtain a second pulse feature, wherein the second pulse feature includes autocorrelation features and entropy features; determining a target state of the target object based on the second pulse feature, and displaying the target state on the screen.

[0130] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence; and extracting entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0131] In one embodiment, when the processor executes the computer program, it further performs the following steps: traversing and extracting the first interval time and the corresponding second interval time in the pulse interval sequence, wherein the lag order of the first interval time and the second interval time is greater than 1; calculating the autocorrelation coefficient of each group of the first interval time and the second interval time; and calculating the autocorrelation characteristics of the pulse interval sequence based on the autocorrelation coefficient of each group.

[0132] In one embodiment, when the processor executes the computer program, it further performs the following steps: calculating the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy class feature, and the weight corresponding to the entropy class feature; and determining the target state based on the target state value.

[0133] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting multiple time-domain features from the pulse interval sequence and extracting multiple nonlinear features from the pulse interval sequence; determining the time-domain state value corresponding to each time-domain feature based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature; determining the nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature; determining a first state value based on each time-domain state value and each nonlinear state value, and determining a first state based on the first state value.

[0134] In one embodiment, when the processor executes the computer program, it further performs the following steps: grouping pulse data according to a preset sliding window to obtain multiple pulse data groups; filtering candidate pulse data in each pulse data group according to preset filtering conditions to obtain multiple candidate pulse peak data; determining the time interval between adjacent candidate pulse peak data, and determining a pulse interval sequence based on each time interval.

[0135] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: determining the pulse interval sequence of the detected object based on pulse data of the detected object within a target time period; performing a first feature extraction on the pulse interval sequence to obtain a first pulse feature, and determining a first state of the detected object based on the first pulse feature, wherein the first pulse feature includes time-domain features and nonlinear features; if the first state is an irregular pulse state, performing a second feature extraction on the pulse interval sequence to obtain a second pulse feature, wherein the second pulse feature includes autocorrelation features and entropy features; determining the target state of the detected object based on the second pulse feature, and displaying the target state on the screen.

[0136] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence; and extracting entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0137] In one embodiment, when the processor executes the computer program, it further performs the following steps: traversing and extracting the first interval time and the corresponding second interval time in the pulse interval sequence, wherein the lag order of the first interval time and the second interval time is greater than 1; calculating the autocorrelation coefficient of each group of the first interval time and the second interval time; and calculating the autocorrelation characteristics of the pulse interval sequence based on the autocorrelation coefficient of each group.

[0138] In one embodiment, when the processor executes the computer program, it further performs the following steps: calculating the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy class feature, and the weight corresponding to the entropy class feature; and determining the target state based on the target state value.

[0139] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting multiple time-domain features from the pulse interval sequence and extracting multiple nonlinear features from the pulse interval sequence; determining the time-domain state value corresponding to each time-domain feature based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature; determining the nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature; determining a first state value based on each time-domain state value and each nonlinear state value, and determining a first state based on the first state value.

[0140] In one embodiment, when the processor executes the computer program, it further performs the following steps: grouping pulse data according to a preset sliding window to obtain multiple pulse data groups; filtering candidate pulse data in each pulse data group according to preset filtering conditions to obtain multiple candidate pulse peak data; determining the time interval between adjacent candidate pulse peak data, and determining a pulse interval sequence based on each time interval.

[0141] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps: determining a pulse interval sequence of the detected object based on pulse data of the detected object within a target time period; extracting a first feature from the pulse interval sequence to obtain a first pulse feature, and determining a first state of the detected object based on the first pulse feature, the first pulse feature including time-domain features and nonlinear features; if the first state is an irregular pulse state, extracting a second feature from the pulse interval sequence to obtain a second pulse feature, the second pulse feature including autocorrelation features and entropy features; determining a target state of the detected object based on the second pulse feature, and displaying the target state on the screen.

[0142] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting autocorrelation features from each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence; and extracting entropy features from each interval time in the pulse interval sequence to obtain the entropy features of the pulse interval sequence.

[0143] In one embodiment, when the processor executes the computer program, it further performs the following steps: traversing and extracting the first interval time and the corresponding second interval time in the pulse interval sequence, wherein the lag order of the first interval time and the second interval time is greater than 1; calculating the autocorrelation coefficient of each group of the first interval time and the second interval time; and calculating the autocorrelation characteristics of the pulse interval sequence based on the autocorrelation coefficient of each group.

[0144] In one embodiment, when the processor executes the computer program, it further performs the following steps: calculating the target state value of the detected object based on the autocorrelation feature, the weight corresponding to the autocorrelation feature, the entropy class feature, and the weight corresponding to the entropy class feature; and determining the target state based on the target state value.

[0145] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting multiple time-domain features from the pulse interval sequence and extracting multiple nonlinear features from the pulse interval sequence; determining the time-domain state value corresponding to each time-domain feature based on the relationship between the first feature value and the corresponding first threshold of each time-domain feature, and the weight of each time-domain feature; determining the nonlinear state value corresponding to each nonlinear feature based on the relationship between the second feature value and the corresponding second threshold of each nonlinear feature, and the weight of each nonlinear feature; determining a first state value based on each time-domain state value and each nonlinear state value, and determining a first state based on the first state value.

[0146] In one embodiment, when the processor executes the computer program, it further performs the following steps: grouping pulse data according to a preset sliding window to obtain multiple pulse data groups; filtering candidate pulse data in each pulse data group according to preset filtering conditions to obtain multiple candidate pulse peak data; determining the time interval between adjacent candidate pulse peak data, and determining a pulse interval sequence based on each time interval.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, pulse data, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0148] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0149] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0150] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of detecting a state of an object, characterized by, The method is applied to a blood pressure measuring instrument, and comprises the following steps: According to the pulse data of the detection object in a target time period, the pulse interval sequence of the detection object is determined, and the target time period is the time for the blood pressure measuring instrument to perform one blood pressure measurement; First feature extraction is performed on the pulse interval sequence to obtain first pulse features, and a first state of the detection object is determined according to the first pulse features, wherein the first pulse features include time domain features and nonlinear features; In the case that the first state is an irregular pulse state, autocorrelation feature extraction is performed on each interval time in the pulse interval sequence to obtain autocorrelation features of the pulse interval sequence, and entropy feature extraction is performed on each interval time in the pulse interval sequence to obtain entropy features of the pulse interval sequence; The target state value of the detection object is calculated according to the autocorrelation features, the weights corresponding to the autocorrelation features, the entropy features and the weights corresponding to the entropy features; The target state is determined according to the target state value, and the target state is displayed on a screen, wherein the target state is an irregular pulse state or a normal state.

2. The method of claim 1, wherein, The autocorrelation feature extraction on each interval time in the pulse interval sequence to obtain the autocorrelation features of the pulse interval sequence comprises the following steps: The first interval time and the corresponding second interval time in the pulse interval sequence are extracted by iteration, and the lag order of the first interval time and the second interval time is greater than 1; The autocorrelation coefficients of each group of the first interval time and the second interval time are calculated; The autocorrelation features of the pulse interval sequence are calculated according to the autocorrelation coefficients of each group.

3. The method of claim 1, wherein, The first feature extraction on the pulse interval sequence to obtain the first pulse features, and the determination of the first state of the detection object according to the first pulse features, comprise the following steps: Time domain feature extraction is performed on the pulse interval sequence to obtain a plurality of time domain features, and nonlinear feature extraction is performed on the pulse interval sequence to obtain a plurality of nonlinear features; The time domain state value corresponding to each time domain feature is determined according to the size relationship between the first feature value of each time domain feature and the corresponding first threshold value, and the weight of each time domain feature; The nonlinear state value corresponding to each nonlinear feature is determined according to the size relationship between the second feature value of each nonlinear feature and the corresponding second threshold value, and the weight of each nonlinear feature; The first state value is determined according to the time domain state values and the nonlinear state values, and the first state is determined according to the first state value.

4. The method according to any one of claims 1 to 3, characterized in that, The determination of the pulse interval sequence of the detection object according to the pulse data of the detection object in a target time period comprises the following steps: The pulse data is grouped according to a preset sliding window to obtain a plurality of pulse data groups; The candidate pulse data in each pulse data group is screened according to a preset screening condition to obtain a plurality of candidate pulse peak data; The time intervals between adjacent candidate pulse peak data are determined, and the pulse interval sequence is determined based on the time intervals.

5. A state detection device of an object, characterized by comprising: The device is applied to a blood pressure measuring instrument, and the device comprises: a pulse interval sequence determination module configured to determine a pulse interval sequence of a detection object according to pulse data of the detection object in a target time period, the target time period being a time period for one blood pressure measurement of the blood pressure measuring instrument; a first feature processing module configured to perform first feature extraction on the pulse interval sequence to obtain first pulse features, and determine a first state of the detection object according to the first pulse features, the first pulse features including time domain features and nonlinear features; a second feature processing module configured to, when the first state is an irregular pulse state, perform autocorrelation feature extraction on each interval time in the pulse interval sequence to obtain autocorrelation features of the pulse interval sequence, and perform entropy feature extraction on each interval time in the pulse interval sequence to obtain entropy features of the pulse interval sequence; a state display module configured to calculate a target state value of the detection object according to a weighted sum of the autocorrelation features, weights corresponding to the autocorrelation features, the entropy features, and weights corresponding to the entropy features, determine a target state according to the target state value, and display the target state on a screen, the target state being an irregular pulse state or a normal state.

6. A sphygmomanometer comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4. The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

Images