CN111631698A - Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint - Google Patents
Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint Download PDFInfo
- Publication number
- CN111631698A CN111631698A CN202010395959.5A CN202010395959A CN111631698A CN 111631698 A CN111631698 A CN 111631698A CN 202010395959 A CN202010395959 A CN 202010395959A CN 111631698 A CN111631698 A CN 111631698A
- Authority
- CN
- China
- Prior art keywords
- blood pressure
- measured
- pressure value
- neural network
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/021—Measuring pressure in heart or blood vessels
- A61B5/02108—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1116—Determining posture transitions
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1118—Determining activity level
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/681—Wristwatch-type devices
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7221—Determining signal validity, reliability or quality
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Abstract
The invention discloses a wearable blood pressure monitoring and correcting method based on motion mode cascade constraint, which comprises the steps of recording information of personnel wearing a photoelectric sensor type intelligent bracelet, data of a nine-axis micro inertial sensor, diastolic pressure and systolic pressure measured by the photoelectric sensor type intelligent bracelet and reference diastolic pressure and systolic pressure measured by an upper arm type electronic sphygmomanometer, and establishing a training sample database. Aiming at the problem that when blood pressure estimation is carried out on the basis of photoelectric blood volume waves in real time, different motion states can interfere blood pressure signals and cause deviation, a cascade classification algorithm is adopted to establish a BP neural network model which considers time between personal basic information, motion states, wearable collected blood pressure values and correction of reference blood pressure values. By the method, the blood pressure value monitored by the photoelectric intelligent bracelet can be corrected, the accuracy of blood pressure measurement is improved, and the exercise state, the exercise amount and the health state based on historical information of a user are judged.
Description
Technical Field
The invention relates to a motion monitoring technology, and belongs to the field of physiological signal monitoring.
Background
Blood pressure is an important parameter reflecting human health conditions, blood pressure examination is a large way for clinically judging diseases and observing treatment effects, and blood pressure monitoring has important significance. At present, the rejuvenation, generalization and complication of hypertension patients with a plurality of common complications and the influence on cardiovascular risks become health problems of the world concern. Wearable equipment's of intelligence function strengthens increasingly, has appeared the bracelet product that can measure blood pressure on the market now, can carry out the measurement of referential to vital sign signals such as user's blood pressure, and the user makes simple judgement to self health condition according to measured data.
At present, an intelligent bracelet for measuring blood pressure generally adopts a photoelectric sensor and a photoelectric sensor, and is assisted by an electrocardiosignal and an oscillometric method. The photoelectric sensor type bracelet can collect the pulse wave of wrist department to carry out analysis to it, and then estimate the blood pressure value. The method is simple and convenient to measure, and most of the smart bracelets on the market adopt the method; the photoelectric sensor is assisted with electrocardiosignals, and the collected electrocardiosignals are added on the basis of the photoelectric sensor for comprehensive analysis. The acquisition mode is complex and the cost is high; the oscillography is similar to the principle of an electronic sphygmomanometer commonly used by people, the blood pressure value is judged through pulse, the measurement is relatively accurate, but the cost is high, and the product volume is large.
At present, a general bracelet adopts a photoelectric sensor to measure blood pressure, namely, blood pressure estimation is carried out based on a blood volume wave (PPG) signal. Most of intelligent bracelets are in poor contact with limbs due to movement in the wearing process, so that PPG photoelectric signal measurement data is inaccurate, blood pressure estimation has large errors and cannot reach medical standards, and the intelligent bracelets cannot be used as medical-grade reference for judging health conditions, diseases and the like. Therefore, based on the motion cascade mode identification and constraint, the error of wearable blood pressure measurement at any time and any place can be improved.
Disclosure of Invention
Based on the problems, the wearable blood pressure monitoring and correcting method based on the movement mode cascade constraint can correct the blood pressure value monitored by the intelligent bracelet based on the photoelectric sensor according to the judged movement state and other information of the user, improve the accuracy of photoelectric blood pressure monitoring and realize monitoring and calibration of the personal dynamic blood pressure within 24 hours.
A wearable blood pressure monitoring method based on motion mode cascade constraint specifically comprises the following steps:
acquiring posture information and motion information of a wearer within a certain time through a nine-axis micro inertial sensor;
establishing a motion mode of the wearer in the current state according to the posture information and the motion information;
measuring diastolic pressure and systolic pressure of a wearer through a smart band based on a photoelectric sensor;
the diastolic pressure and the systolic pressure of the wearer are measured by the upper arm type electronic sphygmomanometer.
A wearable blood pressure monitoring value correction method based on motion mode cascade constraint comprises the following specific steps:
step 1, establishing a sample database according to the measurement data obtained by the method of claim 1: the sample database comprises movement modes of a wearer, basic information, a measured blood pressure value, a reference blood pressure value and a blood pressure value correction quantity, wherein the movement modes comprise six types of standing, sitting still, lying down, walking and running, the basic information comprises height, weight, sex and age, the measured blood pressure value comprises diastolic pressure and systolic pressure measured by an intelligent bracelet, the reference blood pressure value comprises diastolic pressure and systolic pressure measured by an upper arm type electronic sphygmomanometer, and the blood pressure value correction quantity comprises a difference value between the diastolic pressure measured by the intelligent bracelet and the diastolic pressure measured by the upper arm type electronic sphygmomanometer, a difference value between the systolic pressure measured by the intelligent bracelet and the systolic pressure measured by the upper arm type electronic sphygmomanometer;
step 2, training the BP neural network model based on the sample database, and specifically comprising the following steps:
2.1, determining a topological structure of the BP neural network model;
2.2, training the BP neural network model by taking the movement mode, the basic information, the measured blood pressure value and the reference blood pressure value of the wearer in the sample database as the input quantity of the BP neural network model, taking the blood pressure value correction quantity as the output quantity of the BP neural network model, taking a Sigmoid function as a transfer function, taking a gradient descent self-adaptive learning rate algorithm as a training function, taking a threshold learning function as a learning function, and taking the average value and the standard deviation of training sample errors as measurement standards;
and 2.3, obtaining the blood pressure value correction of the testee according to the trained BP neural network model, and outputting the algebraic sum of the measured blood pressure value of the testee and the blood pressure value correction as a corrected blood pressure value.
Advantageous effects
The invention can correct the blood pressure value measured by the photoelectric sensor type intelligent bracelet according to the motion state and the basic information characteristics of personnel. Furthermore, through a motion amount estimation model, an expert health estimation experience model and the like, motion amount statistics and health assessment can be obtained on the basis of motion analysis and blood pressure calibration.
Drawings
FIG. 1 is a system block diagram of a wearable blood pressure monitoring and correction method based on a motion mode cascade constraint;
FIG. 2 is a flow chart of blood pressure and correction based on a kinematic cascade mode constraint;
FIG. 3 is a model diagram for blood pressure value correction based on BP neural network.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the attached drawings:
the invention provides a wearable blood pressure monitoring and correcting method based on motion mode cascade constraint.
As shown in fig. 1 and 2, the present invention proceeds as follows:
step 1, inputting information of personnel wearing a photoelectric sensor type intelligent bracelet, wherein the information comprises height H, weight W, gender G and age A, and the information is used as personal basic information;
step 2, acquiring the acceleration a, the angular velocity omega and the angle theta of 6 motion modes (standing, sitting still, lying, walking and running) of a person wearing the photoelectric sensor type intelligent bracelet through a nine-axis micro-inertial sensor; the data volume under each type is kept above 2 times of the data dimension of the adopted classification algorithm, so that enough training data are obtained;
step 3, acquiring the motion mode in the step 2, and simultaneously measuring photoplethysmography (PPG) by using the photoelectric sensor type intelligent bracelet, estimating diastolic pressure (DBP) and systolic pressure (SBP), wherein the blood pressure value is used as an observation blood pressure value (namely a blood pressure value to be corrected); the data volume under each type is kept above 2 times of the data dimension of the adopted classification algorithm, so that enough training data are obtained;
step 4, estimating a heart rate value (HR) based on an observed waveform of a photoplethysmography (PPG) while acquiring and estimating blood pressure data in the step 3; the data volume under each type is kept above 2 times of the data dimension of the adopted classification algorithm, so that enough training data are obtained;
step 5, collecting the estimated blood pressure data in step 3, and measuring the corresponding diastolic blood pressure by using an upper arm type electronic sphygmomanometer
And systolic pressure
This blood pressure value is taken as a reference blood pressure value (a blood pressure measurement value with an order of magnitude higher accuracy);
and 6, performing data association based on the observation timestamp t, performing data cleaning and filtering smoothing, and discarding the block combined information if the preprocessed data do not accord with physiological common knowledge: if the pre-processing heart rate is not within 40-200 times/partition, discarding the combined data of the block, and performing primary screening on the next combined data; if the systolic pressure is not within the interval of 80-200mmHg, or the diastolic pressure is not within the interval of 30-150 mmHg, or the difference between the systolic pressure and the diastolic pressure is more than 20mmHg after filtering, discarding the combined data of the block;
step 7, according to the data in the step 6, normalization is performed so as to have universality for all people, data partitioning (1 block of data is obtained in each 10 peak-to-peak periods according to continuous and stable preprocessed PPG time domain signals) feature extraction is performed, and finally effective data is obtained for storage, retrieval and classification identification, wherein the input data mainly comprises:
(1) corresponding observation time t;
(2) height H, weight W, gender G, and age A;
(3) the preprocessed micro-inertia acceleration a ', angular velocity omega ', and the characteristic value (maximum value, minimum value, mean value, skewness and variance) of the angle theta ' correspond to 6 motion mode labels when the characteristic value is used for training data;
(4) the diastolic pressure (DBP ') and the systolic pressure (SBP') of the preprocessed wearable photoelectric sphygmomanometer observed quantity and the characteristic values (maximum value, minimum value, mean value, skewness and variance) of the wearable photoelectric sphygmomanometer observed quantity;
(5) diastolic pressure of upper arm electronic sphygmomanometer observed quantity after pretreatment
And systolic pressure
And its eigenvalues (maximum, minimum, mean, skewness, variance);
(6) difference value delta between preprocessed wearable photoelectric blood pressure observed quantity and blood pressure measured by reference upper arm type electronic sphygmomanometerDBP、ΔSBP；
And 8, taking the data in the step 7, the time t and partial data of the personnel information in the step 1 as training samples, taking the parts of the training samples as test samples, and establishing a sample database of the blood pressure value correction model constrained by the movement mode. Learning classification algorithm by adopting parallel and cascade machinesEstablishing a data model of the personnel basic characteristics, the personnel motion state, the observed blood pressure value, the observed heart rate and the reference blood pressure value, and training and testing by adopting a parallel cascade BP neural network model. Finally, through the established BP neural network model, the correction quantity delta of the diastolic pressure/systolic pressure under certain personnel characteristics and motion states is outputDBP'、ΔSBP'. Observing the blood pressure values of the photoelectric sensor type smart bracelet as DBP and SBP, and obtaining the corrected blood pressure value
The step 8 specifically comprises:
step 801, determining the topological structure of the BP neural network:
determining the number of input layers, output layers, hidden layer layers and the number of neurons contained in each hidden layer of the BP neural network model;
step 802, selecting a person motion state, a person basic characteristic, a measured blood pressure value and a reference blood pressure value as input quantities of a BP neural network model, and establishing a blood pressure value correction sample database by taking a blood pressure value correction quantity as an output quantity; the classifier data sets at each level are shown in the following table:
TABLE 1 first-level classifier, dynamic and static pattern recognition, data set format after feature extraction (61D)
TABLE 2 two-stage classifier, static action recognition, feature extracted dataset format (61D)
| Dimension (d) of | Description of the invention |
| 1～4 | 3-axis acceleration and resultant acceleration maximum numerical characteristics |
| 5～8 | 3-axis acceleration and resultant acceleration minimum numerical characteristics |
| 9～12 | 3-axis acceleration and resultant acceleration mean value numerical characteristics |
| 12～16 | 3-axis acceleration and resultant acceleration skewness numerical characteristics |
| 17～20 | 3-axis acceleration and resultant acceleration variance numerical characteristics |
| 21～24 | 3-axis angular velocity and resultant angular velocity maximum value numerical characteristic |
| 25～28 | Numerical characteristics of 3-axis angular velocity and minimum resultant angular velocity |
| 29～32 | 3-axis angular velocity and resultant angular velocity mean value numerical characteristics |
| 32～36 | 3-axis angular velocity and resultant angular velocity skewness numerical characteristics |
| 37～40 | 3-axis angular velocity and resultant angular velocity variance numerical characteristics |
| 41～44 | 3-axis angle and angle maximum value numerical characteristics |
| 45～48 | 3-axis angle and angle minimum numerical characteristics |
| 49～52 | 3-axis angle and angular velocity mean value numerical characteristics |
| 52～56 | Numerical characteristics of 3-axis angle and angular offset |
| 57～60 | 3-axis angle and angle variance numerical characteristics |
| 61 | And (4) classification label: standing, sitting and lying down |
TABLE 3 two-stage classifier, dynamic action recognition, feature extracted dataset format (61D)
TABLE 4 first-level classifier, dynamic static exercise blood pressure modeling, feature extraction post data set format (61D)
| Dimension (d) of | Description of the invention |
| 1～4 | 3-axis acceleration and resultant acceleration maximum numerical characteristics |
| 5～8 | 3-axis acceleration and resultant acceleration minimum numerical characteristics |
| 9～12 | 3-axis acceleration and resultant acceleration mean value numerical characteristics |
| 12～16 | 3-axis acceleration and resultant acceleration skewness numerical characteristics |
| 17～20 | 3-axis acceleration and resultant acceleration variance numerical characteristics |
| 21～24 | 3-axis angular velocity and resultant angular velocity maximum value numerical characteristic |
| 25～28 | Numerical characteristics of 3-axis angular velocity and minimum resultant angular velocity |
| 29～32 | 3-axis angular velocity and resultant angular velocity mean value numerical characteristics |
| 32～36 | 3-axis angular velocity and resultant angular velocity skewness numerical characteristics |
| 37～40 | 3-axis angular velocity and resultant angular velocity variance numerical characteristics |
| 41～44 | 3-axis angle and angle maximum value numerical characteristics |
| 45～48 | 3-axis angle and angle minimum numerical characteristics |
| 49～52 | 3-axis angle and angular velocity mean value numerical characteristics |
| 52～56 | Numerical characteristics of 3-axis angle and angular offset |
| 57～60 | 3-axis angle and angle variance numerical characteristics |
| 61-62 | And (4) classification label: dynamic, static blood pressure model parameters |
TABLE 5 two-stage classifier, blood pressure error assessment, data set format after feature extraction (62D)
| Dimension (d) of | Description of the invention |
| 1～4 | 3-axis acceleration and resultant acceleration maximum numerical characteristics |
| 5～8 | 3-axis acceleration and resultant acceleration minimum numerical characteristics |
| 9～12 | 3-axis acceleration and resultant acceleration mean value numerical characteristics |
| 12～16 | 3-axis acceleration and resultant acceleration skewness numerical characteristics |
| 17～20 | 3-axis acceleration and resultant acceleration variance numerical characteristics |
| 21～24 | 3-axis angular velocity and resultant angular velocity maximum value numerical characteristic |
| 25～28 | Numerical characteristics of 3-axis angular velocity and minimum resultant angular velocity |
| 29～32 | 3-axis angular velocity and resultant angular velocity mean value numerical characteristics |
| 32～36 | 3-axis angular velocity and resultant angular velocity skewness numerical characteristics |
| 37～40 | 3-axis angular velocity and resultant angular velocity variance numerical characteristics |
| 41～44 | 3-axis angle and angle maximum value numerical characteristics |
| 45～48 | 3-axis angle and angle minimum numerical characteristics |
| 49～52 | 3-axis angle and angular velocity mean value numerical characteristics |
| 52～56 | Numerical characteristics of 3-axis angle and angular offset |
| 57～60 | 3-axis angle and angle variance numerical characteristics |
| 61-62 | And (4) classification label: wearable photoelectric blood pressure observation error correction quantity deltaDBP，ΔSBP |
And 803, taking most data in the training sample database as training samples, taking the rest parts as test samples, adopting a Sigmoid function as a transfer function, adopting a gradient descent self-adaptive learning rate algorithm as a training function, and adopting a threshold learning function as a learning function. The classification algorithm can be a decision tree algorithm, a Gaussian naive Bayes algorithm, a K neighbor algorithm, a BP neural network, a support vector machine and a random forest, and the invention adopts the BP neural network as shown in figure 3.
In the invention, a sigmoid function is used as a transfer function in a hidden layer, and a nonlinear function is used as the transfer function in an output layer. Specifically, a unipolar Sigmoid function is adopted:
wherein, x is an input vector, namely the weight value of the motion state of the person, the basic characteristics of the person, the measured blood pressure value and the reference blood pressure value; parameters (a, b) are optimized by a primary action or multiple training of a blood pressure classifier; (x) is the output vector of the hidden layer, which is used as the input to obtain the output quantity through the output layer, i.e. the blood pressure correction quantity.
Through the technical scheme, the blood pressure values observed and measured by the photoelectric sensor type intelligent bracelet are DBP and SBP, and the corrected blood pressure value is obtained
The invention can correct the blood pressure value measured by the photoelectric sensor type intelligent bracelet according to the motion state and the basic information characteristics of personnel. Furthermore, through a motion amount estimation model, a professional health estimation experience model and the like, motion amount statistics and health assessment can be obtained on the basis of motion analysis and blood pressure calibration.
Claims (2)
1. A wearable blood pressure monitoring method based on motion mode cascade constraint is characterized in that: the method specifically comprises the following steps:
acquiring posture information and motion information of a wearer within a certain time through a nine-axis micro inertial sensor;
establishing a motion mode of the wearer in the current state according to the posture information and the motion information;
measuring diastolic pressure and systolic pressure of a wearer through a smart band based on a photoelectric sensor;
the diastolic pressure and the systolic pressure of the wearer are measured by the upper arm type electronic sphygmomanometer.
2. A wearable blood pressure monitoring value correction method based on motion mode cascade constraint is characterized by comprising the following specific steps:
step 1, establishing a sample database according to the measurement data obtained by the method of claim 1: the sample database comprises a movement mode of a wearer, basic information, a measured blood pressure value, a reference blood pressure value and a blood pressure value correction quantity, wherein the movement mode comprises six types of standing, sitting still, lying down, walking and running, the basic information comprises height, weight, sex and age, the measured blood pressure value comprises diastolic pressure and systolic pressure measured by an intelligent bracelet, the reference blood pressure value comprises diastolic pressure and systolic pressure measured by an upper arm type electronic sphygmomanometer, and the blood pressure value correction quantity comprises a difference value between the diastolic pressure measured by the intelligent bracelet and the diastolic pressure measured by the upper arm type electronic sphygmomanometer, a difference value between the systolic pressure measured by the intelligent bracelet and the systolic pressure measured by the upper arm type electronic sphygmomanometer;
step 2, training the BP neural network model based on the sample database, and specifically comprising the following steps:
2.1, determining a topological structure of the BP neural network model;
2.2, training the BP neural network model by taking the movement mode, the basic information, the measured blood pressure value and the reference blood pressure value of the wearer in the sample database as the input quantity of the BP neural network model, taking the blood pressure value correction quantity as the output quantity of the BP neural network model, taking a Sigmoid function as a transfer function, taking a gradient descent self-adaptive learning rate algorithm as a training function, taking a threshold learning function as a learning function, and taking the average value and the standard deviation of the training sample errors as the measurement standard;
and 2.3, obtaining the blood pressure value correction of the testee according to the trained BP neural network model, and outputting the algebraic sum of the measured blood pressure value of the testee and the blood pressure value correction as a corrected blood pressure value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010395959.5A CN111631698A (en) | 2020-05-12 | 2020-05-12 | Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010395959.5A CN111631698A (en) | 2020-05-12 | 2020-05-12 | Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111631698A true CN111631698A (en) | 2020-09-08 |
Family
ID=72323263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010395959.5A Pending CN111631698A (en) | 2020-05-12 | 2020-05-12 | Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111631698A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112401858A (en) * | 2020-12-17 | 2021-02-26 | 北京智康人人科技有限公司 | Method and device for calibrating heart and brain parameters and storage medium |
| CN113663312A (en) * | 2021-08-16 | 2021-11-19 | 东南大学 | Micro-inertia-based non-apparatus body-building action quality evaluation method |
| CN114052678A (en) * | 2021-11-11 | 2022-02-18 | 珠海格力电器股份有限公司 | Information display method and device |
| CN114052689A (en) * | 2021-12-07 | 2022-02-18 | 山东大学 | Blood pressure monitoring device, storage medium and equipment under motion state |
| CN114530028A (en) * | 2022-02-14 | 2022-05-24 | 大连理工大学 | Campus student intelligent bracelet monitoring system and method based on LoRa communication and federal learning |
| CN114601455A (en) * | 2022-05-12 | 2022-06-10 | 电子科技大学 | Motion recognition method based on two-stage neural network |
| CN115500800A (en) * | 2022-03-25 | 2022-12-23 | 张烁 | Wearable physiological parameter detection system |
| CN115670408A (en) * | 2022-12-28 | 2023-02-03 | 可孚医疗科技股份有限公司 | Blood pressure measuring device, linear model coefficient self-correction method and system thereof, and measuring method |
| CN117037993A (en) * | 2023-10-07 | 2023-11-10 | 深圳市爱保护科技有限公司 | Intelligent blood pressure monitoring management method and system |
-
2020
- 2020-05-12 CN CN202010395959.5A patent/CN111631698A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112401858A (en) * | 2020-12-17 | 2021-02-26 | 北京智康人人科技有限公司 | Method and device for calibrating heart and brain parameters and storage medium |
| CN112401858B (en) * | 2020-12-17 | 2023-06-27 | 北京智康人人科技有限公司 | Heart and brain parameter calibration method, device and storage medium |
| CN113663312A (en) * | 2021-08-16 | 2021-11-19 | 东南大学 | Micro-inertia-based non-apparatus body-building action quality evaluation method |
| CN113663312B (en) * | 2021-08-16 | 2022-05-13 | 东南大学 | Micro-inertia-based non-apparatus body-building action quality evaluation method |
| CN114052678A (en) * | 2021-11-11 | 2022-02-18 | 珠海格力电器股份有限公司 | Information display method and device |
| CN114052689A (en) * | 2021-12-07 | 2022-02-18 | 山东大学 | Blood pressure monitoring device, storage medium and equipment under motion state |
| CN114530028A (en) * | 2022-02-14 | 2022-05-24 | 大连理工大学 | Campus student intelligent bracelet monitoring system and method based on LoRa communication and federal learning |
| CN115500800A (en) * | 2022-03-25 | 2022-12-23 | 张烁 | Wearable physiological parameter detection system |
| CN114601455A (en) * | 2022-05-12 | 2022-06-10 | 电子科技大学 | Motion recognition method based on two-stage neural network |
| CN115670408A (en) * | 2022-12-28 | 2023-02-03 | 可孚医疗科技股份有限公司 | Blood pressure measuring device, linear model coefficient self-correction method and system thereof, and measuring method |
| CN117037993A (en) * | 2023-10-07 | 2023-11-10 | 深圳市爱保护科技有限公司 | Intelligent blood pressure monitoring management method and system |
| CN117037993B (en) * | 2023-10-07 | 2024-01-26 | 深圳市爱保护科技有限公司 | Intelligent blood pressure monitoring management method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111631698A (en) | Wearable blood pressure monitoring and correcting method based on motion mode cascade constraint | |
| EP3773155B1 (en) | System and method for non-invasive determination of blood pressure dip based on trained prediction models | |
| Leitner et al. | Personalized blood pressure estimation using photoplethysmography: A transfer learning approach | |
| CN104138253B (en) | A kind of noinvasive arteriotony method for continuous measuring and equipment | |
| CN110420019B (en) | Deep regression heart rate estimation method for ballistocardiogram signals | |
| CN106821356B (en) | Cloud continuous BP measurement method and system based on Elman neural network | |
| WO2019161609A1 (en) | Method for analyzing multi-parameter monitoring data and multi-parameter monitor | |
| JP7045084B2 (en) | Systems and methods for monitoring vascular health | |
| US11617545B2 (en) | Methods and systems for adaptable presentation of sensor data | |
| US20220183569A1 (en) | Blood Pressure Assessment Using Features Extracted Through Deep Learning | |
| Shao et al. | An optimization study of estimating blood pressure models based on pulse arrival time for continuous monitoring | |
| CN115299899A (en) | Activity recognition and beat-to-beat blood pressure monitoring, analyzing and early warning system based on multiple sensors | |
| CN113907727B (en) | Beat-by-beat blood pressure measurement system and method based on photoplethysmography | |
| CN113662520B (en) | Wearable continuous blood pressure measuring system based on uncertainty quantification strategy | |
| CN111466899A (en) | Continuous blood pressure non-invasive monitoring method based on MPGA-MIV-BP model cluster | |
| Roy et al. | BePCon: A photoplethysmography-based quality-aware continuous beat-to-beat blood pressure measurement technique using deep learning | |
| Chan et al. | Estimating heart rate from seismocardiogram signal using a novel deep dominant frequency regressor and domain adversarial training | |
| CN115770028A (en) | Blood pressure detection method, system, device and storage medium | |
| CN114587310A (en) | Method for realizing invasive blood pressure waveform estimation based on photoplethysmography | |
| WO2021045616A1 (en) | A computer-implemented method and an apparatus for determining body positions and/or sleep stages of a person during sleep. | |
| Miah et al. | Heart Rate Extraction from Photoplethysmography Signal: A Multi Model Machine Learning Approach | |
| KR20220123376A (en) | Methods and systems for determining cardiovascular parameters | |
| Villata | Cuffless blood pressure measurement | |
| US20240000323A1 (en) | Photoplethysmography-based real-time blood pressure monitoring system using convolutional bidirectional short- and long-term memory recurrent neural network, and real-time blood pressure monitoring method using same | |
| Jia et al. | Continuous blood pressure prediction based on hierarchical adaptive algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200908 |