An intelligent blood pressure monitor for identifying arrhythmias

Abstract

The present invention discloses a sphygmomanometer for intelligently identifying arrhythmia, and belongs to the technical field of medical equipment. It includes a cuff-type sphygmomanometer composed of a blood pressure measuring device and an electrode sheet, and the cuff-type sphygmomanometer is electrically connected to an intelligent identification subsystem. The present invention solves the problem that existing equipment cannot take into account heart rhythm detection. By adding an electrode sheet, a reminder is provided to the patient whether to perform arrhythmia detection mode when the patient's heart rate and pulse are abnormal. The patient's heart rhythm condition is obtained by using the electrode sheet, and printed by a micro-printer so as to be fed back to the doctor to assist in drawing an accurate conclusion; a decision tree model is used to analyze the type of arrhythmia to which the patient belongs, so as to inform the patient of the true heart rhythm condition; a neural network model is used to correct the patient's blood pressure value when the patient is in arrhythmia, so as to prevent the arrhythmia from affecting the patient's blood pressure value, so that the patient can understand his or her true blood pressure condition and improve the practicality of the sphygmomanometer.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment, in particular to a sphygmomanometer for intelligently identifying arrhythmia. Background Art

[0002] In recent years, with the improvement of living standards, medical health has received more and more attention from the general public; with the accelerated pace of life, there are more and more patients with hypertension in my country. If drug treatment fails to control the increase in blood pressure as expected, the human body will be in a state of long-term hypertension, which may cause irreversible damage to various organs of the body, such as the brain, heart, kidneys, etc.

[0003] Blood pressure measurement is a simple technical operation for medical workers. At the same time, with the development of portable medical equipment, home sphygmomanometers have gradually entered thousands of households and become an indispensable home medical device for people with hypertension and the elderly.

[0004] However, the existing home sphygmomanometers have relatively simple functions and are only used to detect blood pressure and heart rate and pulse, but cannot take into account the heart rhythm detection function. If the patient finds abnormal heart rate and pulse when detecting them by themselves, it cannot be further verified. When they go to the hospital for testing, since the abnormal state of heart rate and pulse has a certain time limit, it is very likely that the heart rate and pulse have returned to normal while they are waiting to arrive at the hospital, and it is difficult for doctors to make correct conclusions based on the patient's oral description.

[0005] Therefore, the existing demand is not met, and we propose an intelligent arrhythmia recognition blood pressure monitor. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent arrhythmia recognition sphygmomanometer. By adding electrodes, the patient is reminded whether to enter the arrhythmia detection mode when the heart rate and pulse are abnormal. The patient's heart rhythm condition is obtained by the electrodes and printed by a micro-printer so as to be fed back to the doctor to assist in drawing an accurate conclusion; a decision tree model is used to analyze the type of arrhythmia to which the patient belongs so as to inform the patient of the true heart rhythm condition; a neural network model is used to correct the patient's blood pressure value when the patient is in arrhythmia, so as to prevent the arrhythmia from affecting the patient's blood pressure value, thereby enabling the patient to understand his or her true blood pressure condition, improving the practicality of the sphygmomanometer, and solving the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] An intelligent arrhythmia recognition sphygmomanometer, comprising: a cuff-type sphygmomanometer composed of a blood pressure measuring device and an electrode sheet, wherein the upper end surface of the cuff-type sphygmomanometer is provided with a display screen for displaying heart rate and blood pressure, and the cuff-type sphygmomanometer is electrically connected to an intelligent recognition subsystem;

[0009] The intelligent identification subsystem comprises:

[0010] A mode selection unit, used to provide the patient with heart rate or blood pressure detection options, and to display the mode options and mode selection results on a display screen;

[0011] The blood pressure detection unit is used to provide patients with blood pressure and heart rate and pulse detection items, and further provide patients with heart rhythm detection reminders based on the heart rate and pulse detection results;

[0012] The blood pressure correction unit is used to provide the patient with multiple blood pressure monitoring item reminders when the patient's blood pressure test results are abnormal, and take the average value of multiple monitoring as the patient's current blood pressure true value to ensure the accuracy of blood pressure testing;

[0013] The heart rhythm detection unit is used to provide the patient with heart rhythm detection items, and judge the patient's heart rhythm condition according to the heart rhythm detection results, and further provide the patient with heart rhythm monitoring reminders and print the patient's heart rhythm detection report;

[0014] The special correction unit is used to collect blood pressure data of historical patients under different arrhythmias as samples, create a neural network model, import the samples into the model for training and testing, and obtain the true blood pressure value under each arrhythmia type; and when the heart rhythm test result of the current patient is arrhythmia, import the currently detected blood pressure value into the trained neural network model for correction to obtain the true blood pressure value of the current patient;

[0015] The data synchronization unit is used to build a client-exclusive APP, which synchronously transmits the patient's heart rate data and blood pressure data to the client-exclusive APP for storage based on Bluetooth communication technology.

[0016] Furthermore, the mode selection unit includes

[0017] Mode 1 selection module, used to provide patients with initial test item reminders, including: blood pressure, heart rate pulse detection and heart rhythm detection;

[0018] The mode two selection module is used to provide the patient with an arrhythmia detection mode when the patient's heart rate and pulse are in an abnormal state after the patient has undergone initial detection.

[0019] Furthermore, the blood pressure detection unit comprises

[0020] The threshold setting module is used to preset the blood pressure threshold and the heart rate pulse threshold in advance, and compare the real-time values ​​of blood pressure and heart rate pulse with the corresponding thresholds to obtain the detection results of blood pressure and heart rate pulse;

[0021] The early warning prompt module is used to issue an early warning prompt on the display screen when the real-time values ​​of the patient's blood pressure and heart rate and pulse exceed or fall below the corresponding thresholds; and when the heart rate and pulse are in an abnormal state, remind the patient whether to turn on the arrhythmia detection mode.

[0022] Furthermore, the blood pressure correction unit comprises

[0023] The frequency setting module is used to set three monitoring modes in the same period for patients with abnormal blood pressure, and remind patients to take blood pressure tests before each monitoring time point;

[0024] The numerical calculation module is used to calculate the average value of the three blood pressure test results after the patient completes the three monitoring modes, and use the obtained average value as the actual blood pressure value of the current patient and display it on the display screen.

[0025] Furthermore, the heart rhythm detection unit comprises

[0026] The abnormal construction module is used to collect the heart rhythm data under historical arrhythmia and perform discretization processing, use the processed heart rhythm data as samples and build a decision tree model, and define the features of multiple groups of heart rhythm data as corresponding arrhythmia types through the decision tree model;

[0027] The abnormality recognition module is used to import the heart rhythm data of the current patient under arrhythmia into the decision tree model for classification, and output the classification result to obtain the arrhythmia type of the current patient, and then feed the type back to the special correction unit for blood pressure value correction;

[0028] The report printing module is a micro printer, which is arranged on one side of the blood pressure measuring device and is used to print the printing content selected by the patient.

[0029] Furthermore, the special correction unit includes

[0030] The sample collection module is used to collect the real-time blood pressure data of historical patients under different arrhythmias, the actual blood pressure values ​​diagnosed afterwards, and the types of arrhythmias to which they belong, and use them as samples for model training;

[0031] The model building module is used to build a neural network model, import samples into the neural network model for training for the first time, and output the training results; then import samples into the neural network model for testing for the second time, and output the test results to determine the accuracy of the neural network model training results;

[0032] The model usage module is used to import the blood pressure data of the current patient under arrhythmia and the feedback arrhythmia type into the neural network model for testing, and output the test results to obtain the actual blood pressure value of the current patient under arrhythmia.

[0033] Furthermore, the sample collection module includes

[0034] The sample processing module is used to clean and unify the blood pressure data of historical patients, and classify the arrhythmia type of each patient, so that the data of each patient includes the real-time blood pressure data under arrhythmia, the true blood pressure value and the arrhythmia type, and the processed data is used as a sample;

[0035] The sample division module is used to divide the samples into training sets and test sets in sequence. The training set is used to establish the characteristic points of the neural network model, and the data set is used to verify and optimize the training results of the neural network model.

[0036] The sample iteration module is used to continuously collect blood pressure data of new patients with different arrhythmias, continue to process the data as samples, and import them into the neural network model for updating and iteration to ensure the accuracy and comprehensiveness of the neural network model.

[0037] Furthermore, the data synchronization unit comprises

[0038] The device connection module is used to confirm whether the cuff-type blood pressure monitor is currently connected to the client-exclusive APP. If connected, the cuff-type blood pressure monitor and the client-exclusive APP establish a Bluetooth transmission link through Bluetooth communication technology;

[0039] The data export module is used to synchronously transmit the patient's heart rate data and blood pressure data to the client's exclusive APP based on the Bluetooth transmission link;

[0040] The data storage module is used to receive and save the patient's heart rate data and blood pressure data in time series.

[0041] Furthermore, the data storage module includes:

[0042] A first encryption module is used to segment the heart rate data to obtain a number of first sub-data; perform clustering processing on the first sub-data to obtain a number of first classification sets; query a preset collection type-hash algorithm data table according to the collection type of the first classification set, determine the corresponding hash algorithm and perform calculation to obtain a hash value corresponding to the first classification set; determine the first proportion information of each first classification set and the heart rate data; perform weighted calculation according to the hash value and the first proportion information corresponding to each first classification set to obtain a first encrypted hash value, encrypt the heart rate data according to the first encrypted hash value to obtain first encrypted data; divide the first encrypted data into first main body data and first identification data;

[0043] The second encryption module is used to segment the blood pressure data to obtain a plurality of second sub-data; perform clustering processing on the second sub-data to obtain a plurality of second classification sets; query the preset collection type-hash algorithm data table according to the collection type of the second classification set, determine the corresponding hash algorithm and perform calculation to obtain the hash value corresponding to the second classification set; determine the second proportion information of each second classification set and the heart rate data; perform weighted calculation according to the hash value and the second proportion information corresponding to each second classification set to obtain a second encrypted hash value, encrypt the heart rate data according to the second encrypted hash value to obtain second encrypted data; divide the second encrypted data into second main body data and second identification data;

[0044] Compute module for:

[0045] The first main body data and the second main body data are stored in the local space, and a storage data table is generated;

[0046] Sending the stored data table to an external memory and calculating the distribution balance of the data stored in the external memory;

[0047]

[0048] in, The distribution balance of the data stored in the external memory; The number of storage nodes included for the external storage; For the The size of the stored data of each storage node; is the total amount of stored data;

[0049] A judgment module is used to judge whether the distribution balance is within a preset distribution balance range. When it is determined that the distribution balance is within the preset distribution balance range, the external memory performs a consensus check on the first main data and the second main data according to the storage data table, stores the identification data corresponding to the main data that passes the consensus check on the external memory, and generates storage error information for the main data that fails the consensus check.

[0050] Furthermore, it also includes: a determination module, which is used to:

[0051] Counting the movement information of the user when wearing the blood pressure measurement device within a preset time period, and calculating the user's motion parameter representation based on the movement information;

[0052]

[0053] in, is the user's motion parameter representation; is a preset time period; is the variance, ; is the mean, ; is the maximum movement speed of the user; The user's mobile location;

[0054] Calculate the reliability of the measurement results based on the user's motion parameter representation;

[0055]

[0056] in, The reliability of the measurement results; is a natural constant; The effective distance of the blood pressure measuring device; is the distance between the blood pressure measuring device and the electrode; is the user's movement direction, which follows a uniform distribution of (0,2π);

[0057] When it is determined that the reliability is less than the preset reliability threshold, it indicates that the current measurement result is invalid, and a re-measurement instruction message is issued.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The present invention, by adding electrodes to the original sphygmomanometer, can provide the patient with a reminder of whether to enter the arrhythmia detection mode when the patient's heart rate and pulse are abnormal, obtain the patient's heart rhythm status through the electrodes and the display screen, and print the electrocardiogram through a micro-printer, so that when the patient learns that the heart rate and pulse are abnormal, the data results can be obtained in time, so as to facilitate feedback to the doctor, assist the doctor in drawing an accurate conclusion, and improve the practicality of the existing cuff-type sphygmomanometer.

[0060] 2. The present invention enriches the intelligent recognition function of the cuff-type blood pressure monitor by constructing a decision tree model and a neural network model. The decision tree model can be used to further analyze the type of arrhythmia of the patient so as to inform the patient of the current true heart rhythm condition; the neural network model is used to correct the patient's blood pressure value when the patient is in arrhythmia, so as to prevent the arrhythmia from affecting the patient's blood pressure value, thereby enabling the patient to understand his or her true blood pressure condition, further improving the practicality of the existing cuff-type blood pressure monitor. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a structural diagram of the intelligent arrhythmia recognition blood pressure monitor of the present invention;

[0062] Figure 2 This is a composition diagram of the intelligent identification subsystem of the present invention.

[0063] In the figure: 1. Cuff-type blood pressure monitor; 11. Blood pressure measuring device; 12. Electrode sheet; 2. Display screen; 3. Micro printer. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In order to solve the problem that the existing home blood pressure monitors are relatively single-function, only used to detect blood pressure and heart rate and pulse, and cannot take into account the heart rhythm detection function; if the patient's heart rate and pulse are abnormal when they detect it by themselves, it is impossible to further verify; when they go to the hospital for testing, the abnormal state of heart rate and pulse has a certain time limit, and it is very likely that the heart rate and pulse have returned to normal while waiting to arrive at the hospital, making it difficult for doctors to make correct conclusions based on the patient's oral statements. Please refer to the technical requirements for Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0066] An intelligent blood pressure monitor for identifying arrhythmias includes a cuff-type blood pressure monitor 1 composed of a blood pressure measuring device 11 and an electrode sheet 12. The blood pressure measuring device 11 is composed of a host and a cuff. The host includes an air pump, a control circuit and a pipeline. This is a device disclosed in the prior art and is not shown one by one in the drawings of this application. The upper end surface of the blood pressure measuring device 11 cuff-type blood pressure monitor 1 is provided with a display screen 2 for displaying heart rate and blood pressure. The display screen 2 is divided into two parts, one for displaying blood pressure values ​​and heart rate and pulse values, and the other for displaying an electrocardiogram; the cuff-type blood pressure monitor 1 is electrically connected to the intelligent identification subsystem, and the intelligent identification subsystem is used to serve the intelligent options of the cuff-type blood pressure monitor 1 and to optimize the intelligent identification function of the cuff-type blood pressure monitor 1;

[0067] Intelligent identification subsystem, including:

[0068] The mode selection unit is used to provide the patient with a heart rate or blood pressure detection option and display the mode option and the mode selection result on the display screen 2; the mode selection unit includes:

[0069] The mode selection module is used to provide the patient with initial detection item reminders, including: blood pressure, heart rate and pulse detection and heart rhythm detection; in this embodiment, for example, when the patient uses the cuff-type sphygmomanometer 1, the patient selects the initial detection item through the buttons arranged on the upper surface of the device, such as: selecting the conventional blood pressure measurement mode, tying the wristband part to the patient's elbow, selecting the start button to pressurize the wristband through the internal air pump and then release it, so as to obtain the patient's current blood pressure value and heart rate and pulse data, and display them on the display screen 2; if the patient knows that he or she has arrhythmia, he or she can directly select the heart rhythm detection item during the initial selection, by respectively attaching three electrode sheets 12 to the left shoulder, the right shoulder and the anterior area of ​​the heart, selecting the start button to obtain the patient's heart rhythm distribution data through the electrode sheets 12, and displaying it on the display screen 2.

[0070] The mode two selection module is used to provide the patient with an arrhythmia detection mode when the patient's heart rate and pulse are in an abnormal state after the initial detection of the patient; following the above embodiment, for example, after the patient selects the conventional blood pressure measurement mode, it is detected through detection that the patient's current heart rate and pulse are abnormal, such as: the heartbeat is too fast, then the display screen 2 will feedback a reminder whether to perform a heart rhythm detection item, and the patient can choose yes or no according to his or her own needs. After the patient chooses yes, the three electrode sheets 12 are pasted according to the above operation, and then the heart rhythm detection service continues for the patient.

[0071] The blood pressure detection unit is used to provide blood pressure and heart rate and pulse detection items for patients, and further provide heart rhythm detection reminders for patients according to the heart rate and pulse detection results; the blood pressure detection unit includes:

[0072] The threshold setting module is used to preset the blood pressure threshold and the heart rate and pulse threshold in advance, and compare the real-time values ​​of blood pressure and heart rate and pulse with the corresponding thresholds respectively to obtain the detection results of blood pressure and heart rate and pulse; specifically, it is clear through clinical trials that the normal range of human blood pressure should be: systolic pressure range is 90-140mmHg, diastolic pressure range is between 60-90mmHg, normal pulse range is 60-100 times / minute, and a pulse of less than 60 times / minute or more than 100 times / minute may indicate arrhythmia; therefore, the blood pressure threshold is set to: systolic pressure 90-140mmHg, diastolic pressure 60-90mmHg, and heart rate and pulse threshold is: 60-100 times / minute.

[0073] Continuing with the above embodiment, for example, after the patient selects the conventional blood pressure measurement mode, it is detected through detection that the patient's current heart rate and pulse are abnormal, such as: the heartbeat is too fast, the heart rate and pulse are 110 beats / minute, and the heart rate and pulse value at this time are compared with its threshold value, and it is found that the current heart rate and pulse have exceeded the threshold range, indicating that the patient is currently in arrhythmia.

[0074] The early warning prompt module is used to issue an early warning prompt on the display screen 2 when the real-time values ​​of the patient's blood pressure and heart rate and pulse exceed or fall below the corresponding threshold value; and when the heart rate and pulse are in an abnormal state, remind the patient whether to turn on the arrhythmia detection mode; in connection with the above embodiment, such as: detecting that the patient's current heart rate and pulse are 110 beats / min, comparing the heart rate and pulse value with its threshold value, and obtaining that the current heart rate and pulse have exceeded the threshold range, it indicates that the patient is currently in the state of arrhythmia, and the result is fed back to the display screen 2 and the mode two selection module, and the mode two selection module is used to issue a reminder to the patient whether to perform the heart rhythm detection project, and the patient's blood pressure exceeds or falls below the threshold value. Feedback is also performed according to this step.

[0075] The blood pressure correction unit is used to provide the patient with reminders of multiple blood pressure monitoring items when the patient's blood pressure test results are abnormal, and take the average value of multiple monitoring as the patient's current true blood pressure value to ensure the accuracy of blood pressure testing; the blood pressure correction unit includes:

[0076] The frequency setting module is used to set three monitoring modes in the same period for patients with abnormal blood pressure, and remind the patient to take a blood pressure test before each monitoring time point. Specifically, when the patient has abnormal blood pressure, multiple monitoring modes are set for the patient, including: measuring blood pressure once every three minutes, for a total of three times, and calculating the average of the three results as the current blood pressure value. At the same time, if it is determined that the current patient has arrhythmia, the maximum inflation volume of the air pump is automatically adjusted to 200mmHg, and the deflation speed is 1mmHg / s, so as to ensure the accuracy of blood pressure measurement.

[0077] The numerical calculation module is used to calculate the average value of the three blood pressure test results after the patient ends the three monitoring modes, and use the obtained average value as the actual blood pressure value of the current patient, and display it on the display screen 2; following the above embodiment, for example, if the three blood pressure test results are 80, 85, and 82 respectively, and the average value of the three is 82, then 82 is used as the blood pressure test result of this time.

[0078] The heart rhythm detection unit is used to provide the patient with heart rhythm detection items, and judge the patient's heart rhythm status according to the heart rhythm detection results, and further provide the patient with heart rhythm monitoring reminders and print the patient's heart rhythm detection report; the heart rhythm detection unit includes:

[0079] The abnormal construction module is used to collect heart rhythm data under historical arrhythmias and perform discretization processing. The processed heart rhythm data is used as samples to build a decision tree model. The features of multiple groups of heart rhythm data are defined as corresponding arrhythmia types through the decision tree model. Specifically, before building the decision tree model, a large amount of heart rhythm data is collected to enrich the classification of decision tree branches. The tree construction algorithm is only applicable to nominal data, so if the data is numerical, it must be discretized. After processing the data, arrhythmia is selected as the total branch of the decision tree model, and too fast heart rate is selected below its total branch. And bradycardia are selected as two branches, and bradycardia, sinus arrest, sick sinus syndrome, and sinoatrial block (I, II, and III degrees) are selected as feature branches under the branch of slow heart rate. At this time, the patient's heart rate is <60 beats / min; premature beats, sinus tachycardia, atrial tachycardia (atrial flutter, atrial fibrillation), supraventricular tachycardia, and ventricular tachycardia (ventricular flutter, ventricular fibrillation) are selected as feature branches under the branch of fast heart rate. At this time, the patient's heart rate is >100 beats / min, so as to enrich the probability of arrhythmia types covered by the decision tree model and make a more accurate distinction between the types of arrhythmias in patients.

[0080] The abnormality recognition module is used to import the heart rhythm data of the current patient under arrhythmia into the decision tree model for classification, and output the classification result to obtain the arrhythmia type of the current patient, and then feed the type back to the special correction unit for blood pressure value correction; specifically, the patient's current heart rhythm data is imported into the decision tree model for training, and the heart rate, pulse and heart rhythm distribution are analyzed to obtain the arrhythmia type of the patient.

[0081] The report printing module is a micro printer 3, which is arranged on one side of the blood pressure measuring device 11 and is used to print the content selected by the patient; so as to record the patient's electrocardiogram in time and provide the doctor with effective medical data when the patient goes to the hospital, so as to obtain more accurate medical conclusions.

[0082] The special correction unit is used to collect blood pressure data of historical patients under different arrhythmias as samples, create a neural network model, import the samples into the model for training and testing, and obtain the true blood pressure value under each arrhythmia type; and when the heart rhythm test result of the current patient is arrhythmia, import the current detected blood pressure value into the trained neural network model for correction to obtain the true blood pressure value of the current patient; the special correction unit includes:

[0083] The sample collection module is used to collect the real-time blood pressure data of historical patients under different arrhythmias, the actual blood pressure values ​​diagnosed afterwards, and the types of arrhythmias to which they belong, and use them as samples for model training; the sample collection module includes:

[0084] The sample processing module is used to clean and unify the blood pressure data of historical patients, and classify the arrhythmia type of each patient, so that the data of each patient includes real-time blood pressure data under arrhythmia, true blood pressure value and the type of arrhythmia, and the processed data is used as a sample; by cleaning the blood pressure data samples, erroneous data and duplicate data are eliminated to ensure the validity and accuracy of the blood pressure data samples; at the same time, the blood pressure data samples are formatted in a unified format so that the format of the blood pressure data samples conforms to the neural network model, thereby improving the accuracy of the neural network model test.

[0085] The sample division module is used to divide the samples into training sets and test sets in sequence. The training set is used to establish the characteristic points of the neural network model, and the data set is used to verify and optimize the training results of the neural network model.

[0086] The sample iteration module is used to continuously collect blood pressure data of new patients with different arrhythmias, continue to process the data as samples, and import the data into the neural network model for updating and iteration to ensure the accuracy and comprehensiveness of the neural network model; by continuously collecting blood pressure data of new patients, the characteristic points and topological structure of the neural network model are continuously optimized, and then the comprehensiveness of the characteristic points in the neural network model is optimized, further improving the accuracy of the neural network model test.

[0087] The model building module is used to build a neural network model, import samples into the neural network model for the first time for training, and output the training results; then import samples into the neural network model for testing for a second time, and output the test results to establish the accuracy of the neural network model training results; specifically, the blood pressure data of different patients and different arrhythmias and the actual blood pressure values ​​diagnosed afterwards and the corresponding arrhythmia types are imported into the neural network model for training and testing, thereby establishing the actual blood pressure value corresponding to a certain arrhythmia type and current blood pressure value of the patient in the neural network model, so that the patient can know the current arrhythmia type and blood pressure value in the future, and can correct the current blood pressure value through the neural network model to obtain the current actual blood pressure value, so that the patient's blood pressure test results are not affected by arrhythmia, thereby ensuring the accuracy of blood pressure detection.

[0088] The model usage module is used to import the blood pressure data of the current patient under arrhythmia and the feedback arrhythmia type into the neural network model for testing, and output the test results to obtain the actual blood pressure value of the current patient under arrhythmia; as mentioned above, after the patient undergoes blood pressure testing with a heart rhythm detector, the type of arrhythmia to which the patient belongs is learned through the decision tree model, and then the type and current blood pressure value are fed back to the neural network model for testing and correction, thereby outputting the patient's current actual blood pressure value.

[0089] The data synchronization unit is used to build a client-exclusive APP, and based on Bluetooth communication technology, the patient's heart rate data and blood pressure data are synchronously transmitted to the client-exclusive APP for storage; the data synchronization unit includes:

[0090] The device connection module is used to confirm whether the cuff-type blood pressure monitor 1 is currently connected to the client-exclusive APP. If connected, the cuff-type blood pressure monitor 1 and the client-exclusive APP establish a Bluetooth transmission link through Bluetooth communication technology; the data export module is used to synchronously transmit the patient's heart rate data and blood pressure data to the client-exclusive APP based on the Bluetooth transmission link; the data storage module is used to receive and save the patient's heart rate data and blood pressure data in time series; through the Bluetooth connection, the cuff-type blood pressure monitor 1 and the client-exclusive APP realize the data interaction function, thereby facilitating the feedback of the patient's real-time test results to the client-exclusive APP, which is convenient for the patient to store and record.

[0091] The beneficial effects achieved by the above content are as follows: through the above method, when the patient learns that the heart rate and pulse are abnormal, the arrhythmia results can be obtained in time to facilitate feedback to the doctor, assisting the doctor to draw accurate conclusions, thereby improving the practicality of the existing cuff-type blood pressure monitor 1; by constructing a decision tree model and a neural network model to enrich the intelligent recognition function of the cuff-type blood pressure monitor, the practicality of the existing cuff-type blood pressure monitor 1 is further improved.

[0092] Working principle: By adding an electrode sheet 12 to the original cuff-type blood pressure monitor 1, when the patient's heart rate and pulse are abnormal, the patient can be reminded whether to enter the arrhythmia detection mode. The patient obtains the current arrhythmia result through the electrode sheet 12 and the display screen 2, and prints his electrocardiogram through the micro printer 3; the decision tree model can be used to further analyze the type of arrhythmia to which the patient belongs, so as to inform the patient of the current true heart rhythm condition; the neural network model is used to correct the patient's blood pressure value when the patient is in arrhythmia, so as to prevent the arrhythmia from affecting the patient's blood pressure value, so that the patient can understand his true blood pressure condition.

[0093] Furthermore, the data storage module includes:

[0094] A first encryption module is used to segment the heart rate data to obtain a number of first sub-data; perform clustering processing on the first sub-data to obtain a number of first classification sets; query a preset collection type-hash algorithm data table according to the collection type of the first classification set, determine the corresponding hash algorithm and perform calculation to obtain a hash value corresponding to the first classification set; determine the first proportion information of each first classification set and the heart rate data; perform weighted calculation according to the hash value and the first proportion information corresponding to each first classification set to obtain a first encrypted hash value, encrypt the heart rate data according to the first encrypted hash value to obtain first encrypted data; divide the first encrypted data into first main body data and first identification data;

[0095] The second encryption module is used to segment the blood pressure data to obtain a plurality of second sub-data; perform clustering processing on the second sub-data to obtain a plurality of second classification sets; query the preset collection type-hash algorithm data table according to the collection type of the second classification set, determine the corresponding hash algorithm and perform calculation to obtain the hash value corresponding to the second classification set; determine the second proportion information of each second classification set and the heart rate data; perform weighted calculation according to the hash value and the second proportion information corresponding to each second classification set to obtain a second encrypted hash value, encrypt the heart rate data according to the second encrypted hash value to obtain second encrypted data; divide the second encrypted data into second main body data and second identification data;

[0096] Compute module for:

[0097] The first main body data and the second main body data are stored in the local space, and a storage data table is generated;

[0098] Sending the stored data table to an external memory and calculating the distribution balance of the data stored in the external memory;

[0099]

[0100] in, The distribution balance of the data stored in the external memory; The number of storage nodes included for the external storage; For the The size of the stored data of each storage node; is the total amount of stored data;

[0101] A judgment module is used to judge whether the distribution balance is within a preset distribution balance range. When it is determined that the distribution balance is within the preset distribution balance range, the external memory performs a consensus check on the first main data and the second main data according to the storage data table, stores the identification data corresponding to the main data that passes the consensus check on the external memory, and generates storage error information for the main data that fails the consensus check.

[0102] The working principle of the above technical solution is as follows: Data segmentation: The heart rate data is segmented into multiple smaller data blocks (first sub-data) for subsequent processing. Clustering processing: The segmented data blocks are clustered to form several data sets with similar characteristics (first classification sets). Hash algorithm selection: According to the type of each classification set, the corresponding hash algorithm is searched and applied from the preset set type-hash algorithm data table to calculate the hash value of each classification set. Proportion information determination: The proportion of each classification set in the original heart rate data is calculated. Weighted calculation: The hash value of each classification set and its proportion information are used to perform weighted calculation to obtain the first encrypted hash value. Data encryption: The original heart rate data is encrypted using the first encrypted hash value to generate the first encrypted data. Data segmentation storage: The encrypted data is further divided into the first main body data and the first identification data for storage and transmission. The first main body data is the main content data, and the first identification data is the index data of the first main body data. The second encryption module is similar to the first encryption module, but processes blood pressure data. It also performs steps such as data segmentation, clustering, hash algorithm selection, proportion information determination, weighted calculation, data encryption and data segmentation storage, but it is aimed at blood pressure data, and generates the second encrypted data and its corresponding second main body data and second identification data. Data storage: The first main body data and the second main body data are stored in the local space, and a storage data table is generated. Data transmission: The storage data table is sent to the external memory. It is convenient to realize the storage of data in different spaces, avoiding the storage of all data in multiple capacities after communication in the prior art. Calculate the distribution balance of the stored data in the external memory to evaluate whether the data is evenly distributed among the storage nodes. The calculation formula takes into account the size of the stored data and the total amount of data of each storage node. When determining unevenness, it is first necessary to adjust the node storage in the external memory so that each storage node can perform effective storage, while ensuring the distribution balance of the stored data in the external memory, realizing the effective storage of the identification data (index data), and facilitating the later data call. According to the distribution balance obtained by the calculation module, it is judged whether it is within the preset distribution balance range. If the distribution balance is within the preset range, the external memory will perform a consensus check on the first main body data and the second main body data according to the storage data table. The identification data corresponding to the main body data that passes the check will be stored on the external memory, and the main body data that fails the check will generate storage error information. It is convenient to adjust the main body data storage in time according to the storage error information, and realize the accuracy of the corresponding relationship between the main body data and the identification data.

[0103] The beneficial effects of the above technical solution are as follows: through sophisticated data processing, encryption and storage strategies, the security, integrity and efficiency of heart rate data and blood pressure data are ensured. At the same time, by calculating the distribution balance of stored data and performing corresponding verification, the reasonable distribution and reliability of data in the external storage are guaranteed. The main data is stored in the local space, and the identification data is stored in the external storage, which improves the efficiency and accuracy of data storage.

[0104] Furthermore, it also includes: a determination module, which is used to:

[0105] Counting movement information of the user when wearing the blood pressure measurement device 11 within a preset time period, and calculating the user's motion parameter representation based on the movement information;

[0106]

[0107] in, is the user's motion parameter representation; is a preset time period; is the variance, ; is the mean, ; is the maximum movement speed of the user; The user's mobile location;

[0108] Calculate the reliability of the measurement results based on the user's motion parameter representation;

[0109]

[0110] in, The reliability of the measurement results; is a natural constant; The effective distance of the blood pressure measuring device; is the distance between the blood pressure measuring device and the electrode; is the user's movement direction, which follows a uniform distribution of (0,2π);

[0111] When it is determined that the reliability is less than the preset reliability threshold, it indicates that the current measurement result is invalid, and a re-measurement instruction message is issued.

[0112] The working principle and beneficial effects of the above technical solution: When measuring, some users may move their arms intentionally or unintentionally, which may lead to inaccurate measurement results. Therefore, when the determination module determines that the user has moved, it judges the reliability of the obtained measurement result. When the reliability is less than the preset reliability threshold, it indicates that the current measurement result is invalid, and a re-measurement instruction message is issued. This facilitates accurate determination of the user's heart rate data and blood pressure data, and improves the accuracy of the measurement.

[0113] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0114] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent blood pressure monitor for identifying arrhythmias, comprising a cuff-type blood pressure monitor (1) consisting of a blood pressure measuring device (11) and an electrode sheet (12), characterized in that The upper end surface of the cuff-type blood pressure monitor (1) is provided with a display screen (2) for displaying heart rate and blood pressure, and the cuff-type blood pressure monitor (1) is electrically connected to the intelligent identification subsystem; The intelligent identification subsystem comprises: A mode selection unit, used to provide the patient with heart rate or blood pressure detection options, and to display the mode options and mode selection results on a display screen (2); The blood pressure detection unit is used to provide patients with blood pressure and heart rate and pulse detection items, and further provide patients with heart rhythm detection reminders based on the heart rate and pulse detection results; The blood pressure correction unit is used to provide the patient with multiple blood pressure monitoring item reminders when the patient's blood pressure test results are abnormal, and take the average value of multiple monitoring as the patient's current blood pressure true value to ensure the accuracy of blood pressure testing; The heart rhythm detection unit is used to provide the patient with heart rhythm detection items, and judge the patient's heart rhythm condition according to the heart rhythm detection results, and further provide the patient with heart rhythm monitoring reminders and print the patient's heart rhythm detection report; The special correction unit is used to collect blood pressure data of historical patients under different arrhythmias as samples, create a neural network model, import the samples into the model for training and testing, and obtain the true blood pressure value under each arrhythmia type; and when the heart rhythm test result of the current patient is arrhythmia, import the currently detected blood pressure value into the trained neural network model for correction to obtain the true blood pressure value of the current patient; The data synchronization unit is used to build a client-exclusive APP, which synchronously transmits the patient's heart rate data and blood pressure data to the client-exclusive APP for storage based on Bluetooth communication technology.

2. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that , the mode selection unit includes Mode 1 selection module, used to provide patients with initial test item reminders, including: blood pressure, heart rate pulse detection and heart rhythm detection; The mode two selection module is used to provide the patient with an arrhythmia detection mode when the patient's heart rate and pulse are in an abnormal state after the patient has undergone initial detection.

3. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that The blood pressure detection unit comprises The threshold setting module is used to preset the blood pressure threshold and the heart rate pulse threshold in advance, and compare the real-time values ​​of blood pressure and heart rate pulse with the corresponding thresholds to obtain the detection results of blood pressure and heart rate pulse; The early warning prompt module is used to issue an early warning prompt on the display screen (2) when the real-time values ​​of the patient's blood pressure and heart rate and pulse exceed or fall below the corresponding threshold values; and to remind the patient whether to turn on the arrhythmia detection mode when the heart rate and pulse are in an abnormal state.

4. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that The blood pressure correction unit comprises The frequency setting module is used to set three monitoring modes in the same period for patients with abnormal blood pressure, and remind patients to take blood pressure tests before each monitoring time point; The numerical calculation module is used to calculate the average value of the three blood pressure test results after the patient completes the three monitoring modes, and use the obtained average value as the actual blood pressure value of the current patient and display it on the display screen (2).

5. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that The heart rhythm detection unit comprises The abnormal construction module is used to collect the heart rhythm data under historical arrhythmia and perform discretization processing, use the processed heart rhythm data as samples and build a decision tree model, and define the features of multiple groups of heart rhythm data as corresponding arrhythmia types through the decision tree model; The abnormality recognition module is used to import the heart rhythm data of the current patient under arrhythmia into the decision tree model for classification, and output the classification result to obtain the arrhythmia type of the current patient, and then feed the type back to the special correction unit for blood pressure value correction; The report printing module is a micro printer (3) which is arranged on one side of the blood pressure measuring device (11) and is used to print the print content selected by the patient.

6. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that ,The special correction unit includes The sample collection module is used to collect the real-time blood pressure data of historical patients under different arrhythmias, the actual blood pressure values ​​diagnosed afterwards, and the types of arrhythmias to which they belong, and use them as samples for model training; The sample division module is used to divide the samples into training sets and test sets in sequence. The training set is used to establish the characteristic points of the neural network model, and the test set is used to verify and optimize the training results of the neural network model. The model building module is used to build a neural network model, import the training set in the sample into the neural network model for training, and output the training results; then import the test set in the sample into the neural network model for testing, and output the test results to establish the accuracy of the training results of the neural network model; The model usage module is used to import the blood pressure data of the current patient under arrhythmia and the feedback arrhythmia type into the neural network model for testing, and output the test results to obtain the actual blood pressure value of the current patient under arrhythmia.

7. The intelligent arrhythmia recognition blood pressure monitor according to claim 6, characterized in that ,The sample collection module includes The sample processing module is used to clean and unify the blood pressure data of historical patients, and classify the arrhythmia type of each patient, so that the data of each patient includes the real-time blood pressure data under arrhythmia, the true blood pressure value and the arrhythmia type, and the processed data is used as a sample; The sample iteration module is used to continuously collect blood pressure data of new patients with different arrhythmias, continue to process the data as samples, and import them into the neural network model for updating and iteration to ensure the accuracy and comprehensiveness of the neural network model.

8. The intelligent arrhythmia recognition blood pressure monitor according to claim 1, characterized in that ,The data synchronization unit includes A device connection module, used to confirm whether the cuff-type blood pressure monitor (1) is currently connected to the client-specific APP, and if connected, to establish a Bluetooth transmission link between the cuff-type blood pressure monitor (1) and the client-specific APP through Bluetooth communication technology; The data export module is used to synchronously transmit the patient's heart rate data and blood pressure data to the client's exclusive APP based on the Bluetooth transmission link; The data storage module is used to receive and save the patient's heart rate data and blood pressure data in time series.

9. The intelligent arrhythmia recognition blood pressure monitor according to claim 8, characterized in that , the data storage module comprises: A first encryption module is used to segment the heart rate data to obtain a number of first sub-data; perform clustering processing on the first sub-data to obtain a number of first classification sets; query a preset collection type-hash algorithm data table according to the collection type of the first classification set, determine the corresponding hash algorithm and perform calculation to obtain a hash value corresponding to the first classification set; determine the first proportion information of each first classification set and the heart rate data; perform weighted calculation according to the hash value and the first proportion information corresponding to each first classification set to obtain a first encrypted hash value, encrypt the heart rate data according to the first encrypted hash value to obtain first encrypted data; divide the first encrypted data into first main body data and first identification data; The second encryption module is used to segment the blood pressure data to obtain a plurality of second sub-data; perform clustering processing on the second sub-data to obtain a plurality of second classification sets; query the preset collection type-hash algorithm data table according to the collection type of the second classification set, determine the corresponding hash algorithm and perform calculation to obtain the hash value corresponding to the second classification set; determine the second proportion information of each second classification set and the heart rate data; perform weighted calculation according to the hash value and the second proportion information corresponding to each second classification set to obtain a second encrypted hash value, encrypt the heart rate data according to the second encrypted hash value to obtain second encrypted data; divide the second encrypted data into second main body data and second identification data; Compute module for: The first main body data and the second main body data are stored in the local space, and a storage data table is generated; Sending the stored data table to an external memory and calculating the distribution balance of the data stored in the external memory; in, The distribution balance of the data stored in the external memory; The number of storage nodes included for the external storage; For the The size of the stored data of each storage node; is the total amount of stored data; A judgment module is used to judge whether the distribution balance is within a preset distribution balance range. When it is determined that the distribution balance is within the preset distribution balance range, the external memory performs a consensus check on the first main data and the second main data according to the storage data table, stores the identification data corresponding to the main data that passes the consensus check on the external memory, and generates storage error information for the main data that fails the consensus check.

Images