Atrial fibrillation detection method and device based on blood pressure measurement pulse oscillation wave characteristic analysis

Abstract

The present invention discloses an atrial fibrillation detection method and device based on the analysis of the pulse oscillation wave characteristics of blood pressure measurement, wherein the device includes a signal acquisition module, an atrial fibrillation diagnosis module, a local storage module, and a display module. The signal acquisition module obtains the pulse oscillation wave in the process of blood pressure measurement. The work of the atrial fibrillation diagnosis module includes the following steps in sequence: filtering and denoising the obtained signal and removing the baseline; extracting the RR interval sequence, and preliminarily locating the atrial fibrillation by the RR interval sequence and the difference; extracting the trough interval sequence and screening out the premature beat signal; multi-feature extraction, including the main wave peak height, the dicrotic wave peak value, and the pulse width at 1 / 5 of the main wave peak height; obtaining multiple one-dimensional arrays based on the extracted features, calculating the corresponding indicators, and obtaining the atrial fibrillation detection results. The local storage module records the pulse oscillation wave and the atrial fibrillation detection results. The display module displays the atrial fibrillation detection results and waveforms in real time. The present invention can detect atrial fibrillation based on the pulse oscillation wave, which has positive significance for the diagnosis and treatment of atrial fibrillation.

Description

Technical Field

[0001] The present invention belongs to the field of signal detection and medical electronic technology, and in particular relates to an atrial fibrillation detection method and device based on blood pressure measurement pulse oscillation wave characteristic analysis. Background Art

[0002] Atrial fibrillation (AF), also known as AF, is a common clinical arrhythmia. The stroke and deterioration of heart function caused by it are important causes of disability and death in patients.

[0003] Clinically, the detection of atrial fibrillation requires patients to go to the hospital for multiple long electrocardiogram examinations. The electrocardiogram equipment is expensive, uncomfortable to wear, and requires doctors to further diagnose the electrocardiogram results. This method is time-consuming and labor-intensive, and is easily affected by "no disease during detection, no detection during disease", which puts a burden on doctors and patients. Wearable devices are portable, easy to operate, comfortable, and cost-saving, which can solve the above problems well. Wearable devices monitor the user's physiological signals in real time in daily life, perform feature analysis on the user's physiological signals, and generate reports on the user's health status, which has gradually become the main choice for patients. At present, there are two types of atrial fibrillation detection algorithms: deep learning and traditional algorithms. Most of the former are based on various neural networks. This algorithm is highly complex and therefore has high requirements on the computing power of the device. It is not currently applicable to wearable devices. Traditional atrial fibrillation detection algorithms mostly use RR interval sequences and related derivative parameter indicators for feature selection. This method is more effective on ECG signals, but unlike ECG signals, pulse oscillation wave signals are affected by the heart's ejection capacity during actual measurement. When atrial fibrillation occurs, the atria vibrate rapidly and irregularly, affecting the heart's ejection function, causing the main wave peak amplitude of the pulse oscillation wave to drop significantly. In some cases, the main wave peak and The fusion of the dicrotic wave of the previous pulse oscillation wave reduces the accuracy of the atrial fibrillation detection method based on the RR interval sequence. Therefore, the atrial fibrillation detection method based on the RR interval sequence cannot ensure the applicability and accuracy on the pulse oscillation wave. Therefore, it is proposed to study an atrial fibrillation detection method and device based on the characteristic analysis of the pulse oscillation wave measured by blood pressure. The pulse oscillation wave signal is measured at the appropriate body part of the patient for an appropriate time to ensure the comfort and ease of operation of atrial fibrillation detection, and then a variety of characteristic parameters are extracted to generate accurate atrial fibrillation detection results, so as to assist patients in the prevention and treatment of atrial fibrillation, and provide doctors with the specific circumstances of atrial fibrillation and diagnostic basis, which has positive significance for the prevention and treatment of atrial fibrillation. Summary of the invention

[0004] Technical issues:

[0005] The present invention aims to solve the problem that the atrial fibrillation detection method based on the RR interval sequence cannot ensure the applicability and accuracy of the pulse oscillation wave, and to achieve the goal of extracting multiple characteristic parameters including the RR interval sequence with a low-complexity algorithm, and detecting atrial fibrillation in real time based on threshold judgment, and finally outputting the atrial fibrillation detection result, so as to assist patients in preventing and treating atrial fibrillation, and provide doctors with the specific circumstances of atrial fibrillation and the basis for diagnosis.

[0006] Technical solution:

[0007] In view of the above defects or improvement needs of the prior art, the present invention provides an atrial fibrillation detection method and device based on the analysis of the pulse oscillation wave characteristics of blood pressure measurement. After the pulse oscillation wave is collected, the collected signal is filtered, denoised, and the baseline is removed; then the atrial fibrillation segment is preliminarily located, and the premature beat signal is screened out, and then the features of the atrial fibrillation segment from which the premature beat signal has been screened out are further extracted, and the corresponding indicators are calculated to obtain accurate atrial fibrillation detection results, and the atrial fibrillation detection results are displayed in real time through a display module.

[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0009] A method for detecting atrial fibrillation based on blood pressure measurement pulse oscillation wave characteristic analysis comprises the following steps:

[0010] S1: During the blood pressure measurement process, a pulse oscillation wave waveform signal is obtained and the pulse oscillation wave signal is preprocessed, including filtering and denoising and removing the baseline;

[0011] S2: Preliminary positioning of atrial fibrillation segments on the processed pulse oscillation wave signal, peak point detection using the processed pulse oscillation wave, thereby obtaining an RR interval sequence, and on this basis calculating the adjacent RR interval difference △RR, calculating the proportion of abnormal value △RR, and combining with threshold judgment to preliminarily position the atrial fibrillation segment;

[0012] S3: Perform the next step of feature point detection on the initially located atrial fibrillation segment, extract the trough interval sequence, calculate the difference between adjacent trough intervals, screen out premature beat signals based on the number of consecutive abnormal heart beats in the trough interval sequence, and narrow the range of the atrial fibrillation segment;

[0013] S4: extract multiple features from the signal that screens out premature beats, including the main wave peak height, the dicrotic wave peak value, and the pulse width at 1 / 5 of the main wave peak height; calculate the corresponding indicators based on multiple one-dimensional arrays obtained from all the extracted features, and combine the threshold judgment to obtain the atrial fibrillation detection result. The specific threshold can be selected according to the accuracy requirement;

[0014] S5: Store the pulse oscillation wave and atrial fibrillation detection results in the local storage module and display them in real time on the display module, including the number of atrial fibrillation occurrences, the occurrence time of atrial fibrillation, the duration of each episode, etc. Combine the previously recorded atrial fibrillation detection results to diagnose the type of atrial fibrillation, judge the trend of the atrial fibrillation condition, and give suggestions.

[0015] Further, calculate △RR and convert its time unit. Calculate that △RR is greater than 20 ms and the proportion of abnormal △RR is greater than 75%. Calculate that △RR is greater than 50 ms and the proportion of abnormal △RR is greater than 60% to preliminarily locate the atrial fibrillation segment.

[0016] Further, calculate the difference between adjacent wave trough interval sequences, convert its time unit, and set an abnormal value A. Record the number of heartbeats with the difference between consecutive wave trough interval sequences greater than A ms as B, and set a threshold C. Assign a value to C according to the required accuracy. When the preliminarily located atrial fibrillation segment B < C, it is judged as a premature beat signal and the premature beat signal is screened out; when the preliminarily located atrial fibrillation segment B ≥ C, the signal is retained for the next feature extraction.

[0017] Further, perform the next feature extraction on the preliminarily located atrial fibrillation signal, including the height of the main wave peak, the peak value of the dicrotic wave, the pulse width at 1 / 5 of the height of the main wave peak. Combine the extracted feature wave trough interval sequence and RR interval sequence, which include both the rhythm change and waveform morphology of the pulse oscillation wave, to obtain one-dimensional arrays respectively. For the above X i (i = 1, 2, 3, 4, 5), calculate the following indicators, including: coefficient of variation of successive differences CVSD, coefficient of variation CVNNI, ratio of adjacent interval differences greater than 50 ms to the total interval PNNI_50, short axis length of Poincaré plot SD1; based on threshold judgment, obtain the atrial fibrillation detection results, including the duration and occurrence frequency of atrial fibrillation.

[0018]

[0019] Among them, represents the root mean square value of the interval difference;

[0020] represents the standard deviation of the interval;

[0021] represents the mean of the intervals, and n is the number of intervals;

[0022] represents the interval value, and n is the number of intervals;

[0023] SD1 represents with ΔX i,j as the x-axis abscissa, ΔX i,j+1Draw a graph for the y-axis ordinate; the obtained line Y = -X + 2*Mean_X i The minor axis of the Poincaré map used to quantify the shape of the Poincaré map in the straight line direction.

[0024] One or more embodiments provide an atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave feature analysis, including: a signal acquisition module, an atrial fibrillation diagnosis module, a local storage module, and a display module. The atrial fibrillation diagnosis module includes: a signal preprocessing module, an atrial fibrillation signal preliminary positioning module, a premature beat signal screening module, and an atrial fibrillation signal analysis module.

[0025] The signal acquisition module is used to acquire the pulse oscillation wave during the blood pressure measurement process, and the acquisition method thereof may be an oscillometric method;

[0026] The signal preprocessing module obtains the pulse oscillation wave from the signal acquisition module, and filters and removes noise and baseline drift of the obtained pulse oscillation wave through the signal conditioning circuit and the MCU;

[0027] The atrial fibrillation signal preliminary positioning module uses the pre-processed pulse oscillation wave to perform feature point detection, extracts the RR interval sequence, and preliminarily locates atrial fibrillation based on the RR interval sequence;

[0028] The premature beat signal screening module screens out premature beat signals based on the number of heart beats with abnormal values ​​appearing continuously in the trough interval sequence difference value;

[0029] The atrial fibrillation signal analysis module extracts the main wave peak height, the dicrotic wave peak value, and the pulse width at 1 / 5 of the main wave peak height for the atrial fibrillation segment after the premature beat signal is screened out, and combines the extracted characteristic trough interval sequence and RR interval sequence, which also includes the rhythm changes and waveform morphology of the pulse oscillation wave, to obtain a one-dimensional array Based on these one-dimensional arrays, the coefficient of variation of continuous interval differences CVSD, coefficient of variation CVNNI, ratio of adjacent interval differences greater than 50ms to total intervals PNNI_50, and Poincare plot short axis length SD1 index are calculated; based on threshold judgment, the diagnosis result of atrial fibrillation is obtained;

[0030] The local storage module records the collected pulse oscillation wave signals and atrial fibrillation detection results in detail, and has the characteristics of being able to meet the requirements of long-term recording, large-capacity storage, and convenient data exchange with other devices, assisting patients in the prevention and treatment of atrial fibrillation, and providing doctors with specific conditions and diagnostic basis for atrial fibrillation;

[0031] The display module can display the collected physiological signals in real time with an appropriate length of time, and highlight the atrial fibrillation signal fragments. The corresponding atrial fibrillation detection report is presented in the form of an alarm pop-up window, including the number of atrial fibrillation occurrences, the time of atrial fibrillation occurrence, the duration of each occurrence, etc., combined with the previously recorded atrial fibrillation detection results, the type of atrial fibrillation is diagnosed, and the trend of atrial fibrillation is judged and suggestions are given. It is convenient for patients to know that their hearts are in an unhealthy state at the first time, and to seek medical treatment and treatment in a timely and targeted manner.

[0032] Beneficial effects:

[0033] 1. This method can analyze the pulse oscillation waves obtained during the blood pressure measurement process and obtain the atrial fibrillation detection results in real time. Users only need to wear relevant wearable devices to complete home diagnosis, which saves time and effort and is convenient and fast.

[0034] 2. This method takes into account the impact of premature beats on atrial fibrillation detection results and sets up a premature beat signal screening module to avoid misdiagnosis caused by premature beats.

[0035] 3. In the feature extraction stage of the pulse oscillation wave, based on the characteristic that the pulse oscillation wave signal contains a large amount of physiological information, a variety of features are extracted, including the rhythmic changes and waveform morphology of the pulse oscillation wave. Compared with the atrial fibrillation detection method based only on the RR interval sequence, the accuracy of atrial fibrillation detection can be improved.

[0036] 4. This method only performs further feature extraction on the atrial fibrillation fragments after screening out the premature beat signals, which reduces the amount of computational tasks. The algorithm complexity of feature extraction is low, and the calculation method of indicators is simple, which can effectively avoid the problem of mobile medical equipment being restricted by computing power.

[0037] 5. The atrial fibrillation test results are detailed, with statistics on the number of atrial fibrillation occurrences, occurrence time, duration, frequency, etc., which are easy to read and more meaningful for reference. The locally stored atrial fibrillation pulse oscillation waves and atrial fibrillation test results can provide a basis for doctors' diagnosis and facilitate targeted treatment for patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Flow chart of the atrial fibrillation detection method based on pulse oscillation wave characteristic analysis in the present invention.

[0039] Figure 2 An example diagram of an atrial fibrillation detection device for pulse oscillation wave characteristic analysis in the present invention.

[0040] Figure 3 A schematic diagram of the structure of the pulse oscillation wave atrial fibrillation diagnosis module in the present invention.

[0041] Figure 4 Characteristic diagram of the pulse oscillation wave in the present invention.

[0042] Figure 5 The characteristic index distribution box diagram of the pulse oscillation wave trough interval sequence in the present invention.

[0043] Figure 6 Flow chart of the pulse oscillation wave atrial fibrillation signal analysis module in the present invention.

[0044] Figure 7 An example diagram of the pulse oscillation wave atrial fibrillation detection results in the present invention. Specific implementation methods

[0045] The present invention is further described in detail below in conjunction with the accompanying drawings and implementation examples. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0046] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0047] Example 1

[0048] This embodiment provides an atrial fibrillation detection method based on blood pressure measurement pulse oscillation wave feature analysis. Figure 1 As shown, the following steps are included:

[0049] Step 1: During the blood pressure measurement process, a pulse oscillation wave waveform signal is obtained, which comes from a wearable blood pressure measurement device and various other devices, including a non-contact blood pressure detection device; then the pulse oscillation wave signal is preprocessed, including filtering and denoising, and removing the baseline;

[0050] Step 2: Perform preliminary positioning of atrial fibrillation segments on the processed pulse oscillation wave signal, perform peak point detection with the processed pulse oscillation wave, thereby obtaining an RR interval sequence, and on this basis calculate the adjacent RR interval difference △RR, calculate the proportion of abnormal value △RR, and preliminarily locate the atrial fibrillation segment; specifically, calculate △RR greater than 20ms, and the proportion of abnormal value △RR is greater than 75%, calculate △RR greater than 50ms, and the proportion of abnormal value △RR is greater than 60%, and preliminarily locate the atrial fibrillation segment;

[0051] Step 3: Perform the next step of feature point detection on the initially located atrial fibrillation segment, extract the trough interval sequence, calculate the difference between adjacent trough intervals, screen out premature beat signals based on the number of consecutive abnormal heart beats in the trough interval sequence, and narrow the range of the atrial fibrillation segment;

[0052] Step 4: Extract multiple features from the signal that screens out premature beats, including the main wave peak height, the dicrotic wave peak value, and the pulse width at 1 / 5 of the main wave peak height; calculate the corresponding indicators based on multiple one-dimensional arrays obtained from all extracted features, including: continuous interval difference variation coefficient CVSD, coefficient of variation CVNNI, ratio of adjacent interval differences greater than 50ms to total intervals PNNI_50, and Poincare map short axis length SD1; based on the calculated values ​​of the four indicators of each feature quantity, combine the threshold judgment to obtain the atrial fibrillation detection result, and the specific threshold can be selected according to the accuracy requirement;

[0053] Step 5: Store the pulse oscillation wave and atrial fibrillation detection results in the local storage module and display them in real time on the display module, including the number of atrial fibrillation occurrences, the time of atrial fibrillation occurrence, the duration of each occurrence, etc. Combined with the previously recorded atrial fibrillation detection results, diagnose the type of atrial fibrillation, judge the trend of atrial fibrillation, and give suggestions.

[0054] Example 2

[0055] This embodiment provides an atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave feature analysis, such as Figure 2 As shown, including:

[0056] The signal acquisition module 1 includes an inflatable cuff 11 and a gas pressure sensor 12; it is used to obtain the pulse oscillation wave during the blood pressure measurement process, with a sampling frequency of 200 Hz and each acquisition time lasting 60 seconds;

[0057] The atrial fibrillation diagnosis module 2 obtains the signal from the signal acquisition module 1, performs atrial fibrillation diagnosis on the acquired pulse oscillation wave, and can be built into a mobile device or other terminal; it includes: a signal preprocessing module 21, an atrial fibrillation signal preliminary positioning module 22, a premature beat signal screening module 23, and an atrial fibrillation signal analysis module 24. These modules are connected in sequence, such as Figure 3 As shown;

[0058] The signal preprocessing module 21 obtains the pulse oscillation wave from the signal acquisition module, and performs filtering, denoising and baseline removal on the obtained pulse oscillation wave;

[0059] The atrial fibrillation signal preliminary positioning module 22 uses the pre-processed pulse oscillation wave to perform feature point detection, extracts the RR interval sequence, and calculates the adjacent RR interval difference △RR on this basis, and calculates the proportion of abnormal values ​​△RR, specifically, the calculated △RR is greater than 20ms, and the proportion of abnormal values ​​△RR is greater than 75%; the calculated △RR is greater than 50ms, and the proportion of abnormal values ​​△RR is greater than 60%, and the atrial fibrillation segment is preliminarily positioned;

[0060] The premature beat signal screening module 23 extracts the trough interval sequence, calculates the difference between adjacent trough interval sequences, and converts it into time units, the unit being milliseconds (ms). Set the abnormal value to 50ms, record the number of heartbeats with a difference between consecutive adjacent trough interval sequences greater than 50ms as B, and set the threshold number of heartbeats to 6. When the atrial fibrillation segment B is less than 6 when initially located, it is judged to be a premature beat signal and the premature beat signal is screened out; when the atrial fibrillation segment B is greater than or equal to 6 when initially located, the signal is retained, and the next step of feature parameter extraction is performed to narrow the range of the atrial fibrillation segment;

[0061] The atrial fibrillation signal analysis module 24 extracts multiple features from the atrial fibrillation segment after the premature beat signal is screened out, such as Figure 4 As shown, it includes the main wave peak height 241, the dicrotic wave peak value 242, the pulse width 243 at 1 / 5 of the main wave peak height, and is combined with the extracted characteristic trough interval sequence 244 and the RR interval sequence 245, and also includes the rhythmic changes and waveform morphology of the pulse oscillation wave, and obtains a one-dimensional array respectively. Based on these one-dimensional arrays, the following indicators are calculated, including: coefficient of variation of consecutive interval differences CVSD, coefficient of variation CVNNI, ratio of adjacent interval differences greater than 50ms to total intervals PNNI_50, and Poincare plot short axis length SD1 indicator; Figure 5 The box plots of the distribution of these four indicators derived from the trough interval sequence characteristics are given. The corresponding thresholds are set for the calculation results of these indicators, such as Figure 6 As shown, when the calculation result exceeds the threshold, the corresponding atrial fibrillation result 1 / 2 / 3 / 4 is output; when the calculation result does not exceed the threshold, the corresponding calculation result is directly discarded;

[0062] The local storage module 3 records the collected pulse oscillation wave signals and atrial fibrillation detection results in detail, and has the characteristics of being able to meet the requirements of long-term recording, large-capacity storage, and convenient data exchange with other devices, assisting patients in the prevention and treatment of atrial fibrillation, and providing doctors with specific conditions and diagnostic basis for atrial fibrillation;

[0063] The display module 4 can display the collected physiological signals in real time with an appropriate length of time, and highlight the atrial fibrillation signal fragments. The corresponding atrial fibrillation detection report is presented in the form of an alarm pop-up window, including the number of atrial fibrillation occurrences, the time of atrial fibrillation occurrence, the duration of each occurrence, etc., combined with the previously recorded atrial fibrillation detection results, the type of atrial fibrillation is diagnosed, and the trend of atrial fibrillation is judged, and suggestions are given, such as Figure 7 As shown; it is convenient for patients to know that their heart is in an unhealthy state at the first time, and to seek medical treatment and treatment in a timely and targeted manner.

[0064] The terms "including" and "having" and any variations thereof of the exemplary embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment comprising a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules that are not listed, or may optionally include other steps or modules that are inherent to these processes, methods, products or equipment.

[0065] The exemplary embodiments described above only express several embodiments of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave characteristic analysis, characterized in that: It includes signal acquisition module, atrial fibrillation diagnosis module, local storage module and display module; The atrial fibrillation diagnosis module includes: a signal preprocessing module, an atrial fibrillation signal preliminary positioning module, a premature beat signal screening module, and an atrial fibrillation signal analysis module; the application method of the atrial fibrillation detection device includes: S1: During the blood pressure measurement process, a pulse oscillation wave waveform signal is obtained, and the pulse oscillation wave signal is preprocessed, including: filtering and denoising, and removing the baseline; S2: Preliminary positioning of atrial fibrillation segments on the processed pulse oscillation wave signal, peak point detection using the processed pulse oscillation wave, thereby obtaining an RR interval sequence, and on this basis calculating the adjacent RR interval difference ΔRR, calculating the proportion of abnormal values ​​ΔRR, and preliminarily positioning the atrial fibrillation segments in combination with threshold judgment; S3: Perform the next step of feature point detection on the initially located atrial fibrillation segment, extract the trough interval sequence, calculate the difference between adjacent trough intervals, screen out premature beat signals based on the number of consecutive abnormal heart beats in the trough interval sequence, and narrow the range of the atrial fibrillation segment; S4: extract multiple features from the signal that screens out premature beats, including: main wave peak height, dicrotic wave peak value, and pulse width at 1 / 5 of the main wave peak height; calculate corresponding indicators based on multiple one-dimensional arrays obtained from all extracted features, and combine threshold judgment to obtain atrial fibrillation detection results, and the specific threshold is selected according to the accuracy requirement; S5: The pulse oscillation wave and atrial fibrillation detection results are stored in the local storage module and displayed in real time on the display module, including the number of atrial fibrillation occurrences, the time of atrial fibrillation occurrence, the duration of each occurrence, etc., and the type of atrial fibrillation is diagnosed in combination with the previously recorded atrial fibrillation detection results, and the trend of atrial fibrillation is determined, and suggestions are given; The pulse width at 1 / 5 of the main wave peak height, the trough interval sequence, the RR interval sequence, and the rhythm changes and waveform shape of the pulse oscillation wave are obtained respectively as one-dimensional arrays. Calculate indicators; The indicators include the coefficient of variation of the difference between consecutive intervals CVSD, the coefficient of variation CVNNI, the ratio of the difference between adjacent intervals greater than 50ms to the total interval PNNI_50, and the short axis length SD1 of the Poincare plot; based on the threshold judgment, the atrial fibrillation detection results are obtained, including the duration and frequency of atrial fibrillation; Said in, represents the root mean square value of the interval difference; represents the standard deviation of the interval; represents the mean of the intervals, and n is the number of intervals; represents the interval value, n is the number of intervals; SD1 represents the i,j is the x-axis horizontal coordinate, ΔX i,j+1 Draw a graph for the y-axis ordinate; the obtained line Y = -X + 2*Mean_X i The minor axis of the Poincaré map used to quantify the shape of the Poincaré map in the straight line direction.

2. The atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave characteristic analysis according to claim 1, characterized in that: The pulse oscillation wave signal comes from wearable blood pressure measurement devices, as well as various other forms of devices, including non-contact blood pressure detection devices.

3. The atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave characteristic analysis according to claim 1, characterized in that: The calculated ΔRR is greater than 20ms, and the abnormal value ΔRR accounts for more than 75%; the calculated ΔRR is greater than 50ms, and the abnormal value ΔRR accounts for more than 60%, and the atrial fibrillation segment is preliminarily located.

4. The atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave characteristic analysis according to claim 1, characterized in that: Based on the number of consecutive abnormal values ​​in the difference between the pulse oscillation wave trough interval sequence, premature beat signals are screened out and the range of atrial fibrillation segments is narrowed.

5. The atrial fibrillation detection device based on blood pressure measurement pulse oscillation wave characteristic analysis according to claim 1, characterized in that: The signal acquisition module is used to obtain the pulse oscillation wave during the blood pressure measurement process; The atrial fibrillation diagnosis module includes: a signal preprocessing module, an atrial fibrillation signal preliminary positioning module, a premature beat signal screening module, and an atrial fibrillation signal analysis module; The signal preprocessing module obtains the pulse oscillation wave from the signal acquisition module, and filters and removes the noise and baseline of the obtained pulse oscillation wave; The atrial fibrillation signal preliminary positioning module uses the pre-processed pulse oscillation wave to perform feature point detection and preliminary locates atrial fibrillation based on the RR interval sequence; The premature beat signal screening module screens out premature beat signals and narrows the range of atrial fibrillation segments based on the number of heartbeats with consecutive abnormal values ​​in the difference between trough intervals. The atrial fibrillation signal analysis module extracts the main wave peak height, the dicrotic wave peak value, and the pulse width at 1 / 5 of the main wave peak height for the atrial fibrillation segment after the premature beat signal is screened out; based on multiple one-dimensional arrays obtained from all extracted features, the corresponding indicators are calculated respectively, and the atrial fibrillation detection results are obtained by combining the threshold judgment. The specific threshold is selected according to the accuracy requirements. The local storage module records the collected pulse oscillation wave signals and atrial fibrillation detection results in detail, and has the characteristics of being able to meet the requirements of long-term recording, large-capacity storage, and convenient data exchange with other devices, assisting patients in the prevention and treatment of atrial fibrillation, and providing doctors with specific conditions and diagnostic basis for atrial fibrillation; The display module can display the collected physiological signals in real time at an appropriate length of time, and highlight the atrial fibrillation signal fragments. The corresponding atrial fibrillation detection report is presented in the form of an alarm pop-up window, including the number of atrial fibrillation occurrences, the time of atrial fibrillation occurrence, the duration of each occurrence, etc., combined with the previously recorded atrial fibrillation detection results, the type of atrial fibrillation is diagnosed, and the trend of atrial fibrillation is judged and suggestions are given; it is convenient for patients to know at the first time that their heart is in an unhealthy state, and to seek medical treatment and treatment in a timely and targeted manner.

Images