CN114041764A - Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal - Google Patents
Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal Download PDFInfo
- Publication number
- CN114041764A CN114041764A CN202111535863.5A CN202111535863A CN114041764A CN 114041764 A CN114041764 A CN 114041764A CN 202111535863 A CN202111535863 A CN 202111535863A CN 114041764 A CN114041764 A CN 114041764A
- Authority
- CN
- China
- Prior art keywords
- brain impedance
- impedance
- brain
- features
- blood pressure
- 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
- A61B5/02125—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics of pulse wave propagation time
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/053—Measuring electrical impedance or conductance of a portion of the body
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
- A61B5/352—Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- 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
-
- 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
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
Abstract
A sleeveless blood pressure estimation method based on brain impedance and electrocardiosignals is characterized by comprising the following steps: (1) reading synchronously acquired brain impedance and electrocardiosignals, preprocessing the brain impedance and the electrocardiosignals, and removing baseline drift and high-frequency noise of the brain impedance and the electrocardiosignals, (2) extracting a series of characteristics including pulse conduction time of the preprocessed brain impedance and electrocardiosignals, and (3) sending the extracted series of characteristics into a regression model obtained by training, and outputting estimated values of systolic pressure and diastolic pressure.
Description
Technical Field
The application relates to a method for estimating blood pressure without cuffs based on brain impedance and electrocardiosignals.
Blood pressure is an important physiological indicator of cardiovascular disease. Accurate, continuous blood pressure monitoring is an effective tool for early prevention and treatment of cardiovascular diseases. Conventional blood pressure measurements typically employ an inflatable cuff-based sphygmomanometer. This method requires applying pressure to the subject's body through an inflatable cuff. Such cuff-based blood pressure measurement techniques are uncomfortable for the user and are not suitable for continuous monitoring of blood pressure. Therefore, continuous, non-invasive and long-term cuff-free blood pressure monitoring technology is a research hotspot in recent years.
Brain impedance measurement is a non-invasive, non-invasive method of brain monitoring. The detected brain impedance can be used for representing the cerebral blood flow condition and noninvasive estimation of intracranial pressure, and the risk of invasive detection of intracranial pressure infection such as spinal puncture and craniotomy is avoided. When the intracranial pressure is monitored clinically and non-invasively based on the brain impedance, the blood pressure of a patient is generally monitored, and at the moment, if the cuff-free blood pressure estimation can be carried out based on the brain impedance and the electrocardiosignals instead of the cuff-free blood pressure estimation based on the electrocardiosignals and other signals, one signal can be detected less, the number of sensors attached to the body of the patient is reduced, and the comfort is improved.
Background
Methods for cuff-less blood pressure estimation are mostly based on pulse transit time calculations. The pulse transit time refers to the time interval between the pulse transmissions at two different parts of the human body. Since the pulse starts with the ejection of blood from the heart, the origin of propagation of the pulse is often defined based on the electrocardiographic signal; in addition, the pulse signal of one part is detected to calculate the time interval of propagation. The pulse signal of another part can be directly detected based on pressure sensors such as piezoresistance, piezoelectricity and pressure capacitance, and can also be indirectly detected based on photoelectric sensors according to the absorption condition of light.
Although the detected human body impedance signal does not correspond to the pulse signal of a specific part, the detected human body impedance signal also has the characteristic of pulsation along with heartbeat, because the blood flow volume is periodically changed according to the cardiac cycle of the human body, and compared with the electrocardiosignal, the impedance pulsation has certain hysteresis than the fluctuation of the electrocardiosignal, which is similar to the pulse wave. Therefore, a method of estimating a cuff-less blood pressure based on a detected electrocardiographic signal and a human body impedance signal is also proposed in the literature.
A method for estimating blood pressure without cuff based on electrocardiosignals and human impedance signals relates to the following steps: synchronously acquiring electrocardiosignals of a testee and impedance signals of two points on the body surface, preprocessing to remove baseline drift and high-frequency noise interference, extracting characteristics including conduction time characteristics, heart rate characteristics, various impedance wave morphological characteristics and the like, and finally performing regression prediction to output a blood pressure estimated value based on the extracted characteristics.
In the conventional human impedance detection for estimating the cuff-less blood pressure, some parts for detecting impedance are selected between two upper limbs, some parts are selected between two lower limbs, and some parts are selected as two points on the upper and lower parts of a single limb. In the prior art, no document is available for noninvasive estimation of cuff-free blood pressure based on detection of impedance waveforms and electrocardiosignals of two points of the brain.
Disclosure of Invention
Object of the Invention
The cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignal is provided, and the systolic pressure and the diastolic pressure can be stably and accurately estimated with high robustness.
Technical scheme
A sleeveless blood pressure estimation method based on brain impedance and electrocardiosignals is characterized by comprising the following steps: (1) reading synchronously acquired brain impedance and electrocardiosignals, preprocessing the brain impedance and the electrocardiosignals, and removing baseline drift and high-frequency noise of the brain impedance and the electrocardiosignals, (2) extracting a series of characteristics including pulse conduction time of the preprocessed brain impedance and electrocardiosignals, and (3) sending the extracted series of characteristics into a regression model obtained by training, and outputting estimated values of systolic pressure and diastolic pressure.
According to the sleeveless blood pressure estimation method based on the brain impedance and the electrocardiosignals, the step (2) extracts a series of characteristics including pulse conduction time of the preprocessed brain impedance and the preprocessed electrocardiosignals, and is characterized in that the time when the R peak of the electrocardiosignals appears in the current heartbeat cycle is TRThe time when the maximum value and the minimum value of the brain impedance signal occur is TmaxAnd TminThe maximum value of the brain impedance first-order difference signal occurs at a time TMDThere are three definitions of pulse transit time:
PTTmax=Tmax-TR (1)
PTTmin=Tmin-TR (2)
PAT=TMD-TR (3)
according to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, a series of characteristics including pulse transit time are extracted from the preprocessed brain impedance and electrocardiosignals in the step (2), and the cuff-free blood pressure estimation method is characterized by also including heart rate characteristics besides the pulse transit time.
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, the step (2) extracts a series of characteristics including pulse transmission time from the preprocessed brain impedance and electrocardiosignals, and is characterized in that besides the pulse transmission time, the method also comprises a plurality of morphological characteristics of the brain impedance, and the moments of the maximum value and the minimum value of the brain impedance signals appearing in the current heartbeat cycle are respectively set as TmaxAnd TminThe minimum value of the brain impedance signal in the next heartbeat cycle occurs at the time T'minThen, the width PW, the systolic width SW, and the diastolic width DW of the brain impedance morphological features can be respectively defined as
PW=T′min-Tmin (4)
SW=Tmax-Tmin (5)
DW=T′min-Tmax (6)
PWxCan be defined as the corresponding brain impedance width, SW, when the height of the brain impedance waveform is x% of the maximum valuexCan be defined as the width of the systolic phase, DW, corresponding to when the height of the brain impedance waveform is x% of the maximum valuexCan be defined as the corresponding diastolic width when the height of the brain impedance waveform is x% of the maximum value; PWRxIs defined as PWxRatio to PW, i.e.
Examples hereinafterIn the application, PW and PW are extracted together25,PW50,PW75,PW90,SW,SW25,SW50,SW75,SW90,DW,DW25,DW50,DW75,DW90,PWR25,PWR50,PWR75,PWR90There are 19-dimensional brain impedance morphology features.
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, the preprocessed brain impedance and electrocardiosignals are extracted in the step (2) to obtain a series of characteristics including pulse transit time, and the cuff-free blood pressure estimation method is characterized by further comprising a plurality of height characteristics of the brain impedance, namely the maximum value AM of the brain impedance, besides the pulse transit timemaxMinimum value of brain impedance AMminHeight AM of brain impedance waveform corresponding to the time when brain impedance first-order difference signal is maximumMDThe difference PP between the maximum and minimum values of the brain impedance, and two height ratio features ARmaxAnd ARMDIs defined as
According to the sleeveless blood pressure estimation method based on the brain impedance and the electrocardiosignals, in the step (2), a series of characteristics including pulse transmission time are extracted from the preprocessed brain impedance and the preprocessed electrocardiosignals, and the sleeveless blood pressure estimation method is characterized by comprising slope characteristics of the brain impedance besides the pulse transmission time, and setting the minimum value of the brain impedance signals in the next heartbeat cycle to be AM'minThe brain impedance slope characteristic-the slope AS of the rise period and the slope DS of the fall period of the brain impedance waveform can be defined AS
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, a series of characteristics including pulse transit time are extracted from the preprocessed brain impedance and electrocardiosignals in the step (2), and the cuff-free blood pressure estimation method is characterized by further comprising statistical characteristics of the brain impedance, namely standard deviation SD, skewness Sew and Kurt of the brain impedance, besides the pulse transit time.
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, a series of characteristics including pulse transit time are extracted from the preprocessed brain impedance and electrocardiosignals in the step (2), and the cuff-free blood pressure estimation method is characterized by further comprising sample entropy and approximate entropy characteristics of the brain impedance signals besides the pulse transit time.
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, a series of characteristics including pulse transit time are extracted from the preprocessed brain impedance and electrocardiosignals in the step (2), and the cuff-free blood pressure estimation method is characterized by also comprising morphological characteristics of a brain impedance first-order difference signal besides the pulse transit time. Examples of the latter, including the brain impedance first order difference signal, maximum AMdmaxPulse width PWdPulse width PWd at 50% maximum brain impedance50,PWd50And PWdAnd the slope AS of the rise period of the brain impedance first-order difference signaldAnd a falling period slope DSdTotaling 6-dimensional brain impedance first-order difference morphological characteristics, wherein
According to the cuff-free blood pressure estimation method based on the brain impedance and the electrocardiosignals, the extracted series of characteristics are sent to a regression model obtained through training in the step (3), and estimated values of systolic pressure and diastolic pressure are output.
Advantageous effects
At present, no complete method for estimating the cuff-free blood pressure based on the brain impedance and the electrocardiosignal is reported in the literature.
13 subjects were enrolled in the experiment and in total 1942 valid data were collected on the label of ideal values for systolic and diastolic blood pressure. Dividing the collected data into training data and test data according to the proportion of 7: 3, and evaluating the accuracy of the blood pressure estimation method by using average error (ME), Root Mean Square Error (RMSE) and correlation coefficient (R) for the test data.
In the step (3), the extracted 43-dimensional features are sent to a random forest regression model for blood pressure estimation, and the estimation precision of the systolic pressure is as follows: mean error ME of 0.08mmHg, root of mean square error RMSE of 4.11mmHg, R of 0.90; the accuracy of the estimate of diastolic pressure is: ME was 0.01mmHg, RMSE was 3.36mmHg, and R was 0.88. Therefore, the method can stably and reliably estimate the systolic pressure and the diastolic pressure. The method meets the requirements of the Association for the development of medical devices (AAMI), namely that the average error is less than 5mmHg and the root mean square error is less than 8 mmHg.
The method comprises the following steps (3) of sending the extracted 43-dimensional features into a random forest regression model for blood pressure estimation, wherein the cumulative error of systolic pressure estimation is less than 5mmHg and is 83.1 percent, the cumulative error is less than 10mmHg and is 95.0 percent, and the cumulative error is less than 15mmHg and is 98.6 percent; the ratio of the cumulative error of diastolic pressure estimation of less than 5mmHg was 86.9%, the ratio of the cumulative error of less than 10mmHg was 97.7%, and the ratio of the cumulative error of less than 15mmHg was 99.1%. According to the british standards of the hypertension association (BHS), the systolic and diastolic estimates of the method of the present application are rated on a scale, i.e. the estimated cumulative error is greater than 60% for less than 5mmHg, greater than 85% for less than 10mmHg and greater than 95% for less than 15 mmHg.
Tables 1 and 2 show the performance of the blood pressure estimation of the method of the present application.
TABLE 1 comparison of the accuracy of blood pressure estimation by the method of the present application with AAMI standards
TABLE 2 comparison of the accuracy of blood pressure estimation with the BHS standard in the methods of the present application
Drawings
Fig. 1 is a block diagram of a cuff-less blood pressure estimation method based on brain impedance and electrocardiosignals.
Fig. 2 is a real view of the synchronous measurement of the brain impedance signal and the cardiac signal.
Fig. 3 is a waveform diagram of the brain impedance and the electrocardiosignal before and after baseline drift removal.
Fig. 4 is a schematic diagram of the preprocessed brain impedance signal and the preprocessed electrocardiosignal in the embodiment. The waveforms after filtering the electrocardiosignals and the brain impedance signals are sequentially formed from top to bottom.
Fig. 5 is a schematic diagram of extraction of features of the brain impedance signal and the cardiac signal.
Detailed Description
Examples are given. A brain impedance and electrocardiosignal acquisition system is built, 13 subjects are recruited, the brain impedance and the electrocardiosignals are synchronously detected at a sampling rate of 100kHz, 1942 effective data of ideal values of systolic pressure and diastolic pressure are acquired, and a data measurement live view is shown in figure 2. The present application relates to the algorithmic portion of the system, as described above, implementing the steps as follows:
(1) and reading the synchronously acquired brain impedance and electrocardiosignals, preprocessing the brain impedance and the electrocardiosignals, and removing baseline drift and high-frequency noise of the brain impedance and the electrocardiosignals. In the implementation, a brain impedance signal is picked up from the head of a subject, an electrocardiosignal is picked up from two wrists of the subject, a Savitzky-Golay filter is adopted to remove baseline drift, the polynomial fitting order is 3, the window size is 10001, FIR band-pass filtering of 0.1Hz-50Hz is adopted, and the number of taps is about 2001. The process of removing baseline wander in the original brain impedance signal and the electrocardiosignal is shown in fig. 3, and the signal waveform after removing the baseline wander and the high-frequency noise is shown in fig. 4.
(2) A series of features including pulse transit time are extracted from the preprocessed brain impedance and cardiac electrical signals. In the implementation, the processed brain impedance signal and the processed electrocardiosignal are segmented, and a segment is formed every 8 seconds; then, feature extraction is performed on the signal segment of 8 seconds, and 43-dimensional features including pulse transit time, heart rate features, a plurality of morphological features of brain impedance, a plurality of height features of brain impedance, slope features of brain impedance, statistical features of brain impedance, entropy features of brain impedance, and morphological features of first-order difference signals of brain impedance are extracted, which are shown in table 3. A schematic representation of the meaning of several features is shown in fig. 5. The heart rate, the pulse conduction time, the morphological characteristics, the height characteristics and the slope characteristics of the brain impedance, and the morphological characteristics of the first-order difference signals of the brain impedance are all the characteristics of calculating once per heart cycle, the median is taken as the characteristics of the signal segment of 8 seconds, and the statistical characteristics and the entropy characteristics of the brain impedance are the characteristics of calculating once per signal segment of 8 seconds.
Table 3 43-dimensional features extracted in the examples of the present application
(3) And (4) sending the extracted series of characteristics into a regression model obtained by training, and outputting estimated values of systolic pressure and diastolic pressure. In implementation, the extracted 43-dimensional features are fed into a random forest regression model, and estimated values of systolic pressure and diastolic pressure are output. When training the random forest model, dividing a data set consisting of brain impedance, electrocardio signals, labels of systolic pressure and diastolic pressure after preprocessing into training data and test data according to the proportion of 7: 3, and training two random forest regression models by using the training data for respectively estimating the systolic pressure and the diastolic pressure. The trained random forest regression model contains 500 decision trees with a leaf node having a minimum sample number of 1.
Claims (10)
1. A sleeveless blood pressure estimation method based on brain impedance and electrocardiosignals is characterized by comprising the following steps: (1) reading synchronously acquired brain impedance and electrocardiosignals, preprocessing the brain impedance and the electrocardiosignals, and removing baseline drift and high-frequency noise of the brain impedance and the electrocardiosignals, (2) extracting a series of characteristics including pulse conduction time of the preprocessed brain impedance and electrocardiosignals, and (3) sending the extracted series of characteristics into a regression model obtained by training, and outputting estimated values of systolic pressure and diastolic pressure.
2. The cuff-free blood pressure estimation method based on brain impedance and electrocardiosignal as claimed in claim 1, wherein the step (2) is to extract a series of features including pulse transit time from the preprocessed brain impedance and electrocardiosignal, wherein the time when the R peak of the electrocardiosignal appears in the current heartbeat cycle is TRThe time when the maximum value and the minimum value of the brain impedance signal occur is TmaxAnd TminThe maximum value of the brain impedance first-order difference signal occurs at a time TMDThere are three definitions of pulse transit time:
PTTmax=Tmax-TR (1)
PTTmin=Tmin-TR (2)
PAT=TMD-TR (3)
3. the cuff-free blood pressure estimation method based on brain impedance and cardiac signal as claimed in claim 1, wherein the step (2) extracts a series of features including pulse transit time from the preprocessed brain impedance and cardiac signal, wherein the features include heart rate features in addition to the pulse transit time.
4. The cuff-free blood pressure estimation method based on brain impedance and electrocardiosignal as claimed in claim 1, wherein the step (2) is to extract a series of features including pulse transit time from the preprocessed brain impedance and electrocardiosignal, wherein the features include morphological features of brain impedance in addition to pulse transit time, and the maximum and minimum values of the brain impedance signal in the current heartbeat cycle are respectively set as TmaxAnd TminThe minimum value of the brain impedance signal in the next heartbeat cycle occurs at the time T'minThen, the width PW, the systolic width SW, and the diastolic width DW of the brain impedance morphological features can be respectively defined as
PW=T′min-Tmin (4)
SW=Tmax-Tmin (5)
DW=T′min-Tmax (6)
PWxCan be defined as the corresponding brain impedance width, SW, when the height of the brain impedance waveform is x% of the maximum valuexCan be defined as the width of the systolic phase, DW, corresponding to when the height of the brain impedance waveform is x% of the maximum valuexCan be defined as the corresponding diastolic width when the height of the brain impedance waveform is x% of the maximum value; PWRxIs defined as PWxRatio to PW, i.e.
5. The cuff-less blood pressure estimation method according to claim 1 based on brain impedance and cardiac electrical signalsThe method comprises (2) extracting a series of characteristics including pulse transit time from the preprocessed brain impedance and electrocardiosignal, and is characterized in that the method can also include a plurality of height characteristics of the brain impedance, namely the maximum value AM of the brain impedance, in addition to the pulse transit timemaxMinimum value of brain impedance AMminHeight AM of brain impedance waveform corresponding to the time when brain impedance first-order difference signal is maximumMDThe difference PP between the maximum and minimum values of the brain impedance, and two height ratio features ARmaxAnd ARMDIs defined as
6. The sleeveless blood pressure estimation method based on brain impedance and electrocardiosignals according to claim 1, wherein the step (2) is to extract a series of features including pulse transit time from the preprocessed brain impedance and electrocardiosignals, wherein the features include slope features of the brain impedance in addition to the pulse transit time, and the minimum value of the brain impedance signal in the next heartbeat cycle is set to be AM'minThe slope characteristic of the brain impedance, namely the slope AS of the ascending period and the slope DS of the descending period of the brain impedance waveform, can be respectively defined AS
7. The cuff-free blood pressure estimation method based on brain impedance and electrocardiosignal of claim 1, wherein the extraction of the preprocessed brain impedance and electrocardiosignal in step (2) comprises a series of features of pulse transit time, and the method further comprises the statistical features of brain impedance, namely standard deviation SD, skewness Skaew and Kurt of the brain impedance, besides the pulse transit time.
8. The cuff-free blood pressure estimation method based on brain impedance and electrocardiosignal of claim 1, wherein the extraction of the preprocessed brain impedance and electrocardiosignal in step (2) comprises a series of features of pulse transit time, and the features of sample entropy and approximate entropy of brain impedance can be included besides pulse transit time.
9. The cuff-free blood pressure estimation method based on brain impedance and electrocardiosignal as claimed in claim 1, wherein the step (2) extracts a series of features including pulse transit time from the preprocessed brain impedance and electrocardiosignal, and the features include morphological features of the brain impedance first-order difference signal in addition to the pulse transit time.
10. The sleeveless blood pressure estimation method based on brain impedance and electrocardio signals, as claimed in claim 1, wherein the step (3) is to input the extracted series of features into a regression model obtained by training and output estimated values of systolic pressure and diastolic pressure, and is characterized in that a random forest regression model is adopted, a data set consisting of the brain impedance, the electrocardio signals, the systolic pressure and the diastolic pressure labels after preprocessing in the step (1) is divided into training data and testing data according to a ratio of 7: 3 during model training, two random forest regression models are trained by using the training data, model testing is performed by using the testing data, and finally the training data are respectively used for estimation of the systolic pressure and the diastolic pressure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111535863.5A CN114041764A (en) | 2021-12-15 | 2021-12-15 | Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111535863.5A CN114041764A (en) | 2021-12-15 | 2021-12-15 | Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114041764A true CN114041764A (en) | 2022-02-15 |
Family
ID=80213023
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111535863.5A Pending CN114041764A (en) | 2021-12-15 | 2021-12-15 | Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114041764A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130274615A1 (en) * | 2012-04-12 | 2013-10-17 | Shlomi Ben-Ari | Measurement of Cerebral Physiologic Parameters Using Bioimpedance |
US20140249432A1 (en) * | 2010-12-28 | 2014-09-04 | Matt Banet | Body-worn system for continuous, noninvasive measurement of cardiac output, stroke volume, cardiac power, and blood pressure |
CN104545854A (en) * | 2015-01-30 | 2015-04-29 | 中国科学院电子学研究所 | Cuffless ambulatory blood pressure monitoring equipment based on electrocardio signals and impedance signals |
CN104720777A (en) * | 2015-03-10 | 2015-06-24 | 中国科学院电子学研究所 | Non-invasive continuous blood pressure physiological monitoring system |
US20190307337A1 (en) * | 2014-03-06 | 2019-10-10 | Scanadu Incorporated | Methods and apparatus for self-calibrating non-invasive cuffless blood pressure measurements |
-
2021
- 2021-12-15 CN CN202111535863.5A patent/CN114041764A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140249432A1 (en) * | 2010-12-28 | 2014-09-04 | Matt Banet | Body-worn system for continuous, noninvasive measurement of cardiac output, stroke volume, cardiac power, and blood pressure |
US20130274615A1 (en) * | 2012-04-12 | 2013-10-17 | Shlomi Ben-Ari | Measurement of Cerebral Physiologic Parameters Using Bioimpedance |
US20190307337A1 (en) * | 2014-03-06 | 2019-10-10 | Scanadu Incorporated | Methods and apparatus for self-calibrating non-invasive cuffless blood pressure measurements |
CN104545854A (en) * | 2015-01-30 | 2015-04-29 | 中国科学院电子学研究所 | Cuffless ambulatory blood pressure monitoring equipment based on electrocardio signals and impedance signals |
CN104720777A (en) * | 2015-03-10 | 2015-06-24 | 中国科学院电子学研究所 | Non-invasive continuous blood pressure physiological monitoring system |
Non-Patent Citations (3)
Title |
---|
LIU SH, CHENG DC, SU CH: "A Cuffless Blood Pressure Measurement Based on the Impedance Plethysmography Technique", 《SENSORS (BASEL)》 * |
姬军: "胸—头脉搏波传导时间测量技术研究", 《博士电子期刊》 * |
曹俊魏,程云章: "基于脉搏波信号的无创血压测量的研究进展与展望", 《生物医学工程研究》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106974631B (en) | Blood pressure measuring method and device based on pulse wave waveform and electrocardiosignal | |
Kurylyak et al. | A Neural Network-based method for continuous blood pressure estimation from a PPG signal | |
CN106028918B (en) | Body-worn system for continuous non-invasive measurement of vital signs | |
Ghosh et al. | Continuous blood pressure prediction from pulse transit time using ECG and PPG signals | |
CN106821356B (en) | Cloud continuous BP measurement method and system based on Elman neural network | |
CN107854123B (en) | Sleeveless continuous blood pressure monitoring method and device | |
WO2017024457A1 (en) | Blood-pressure continuous-measurement device, measurement model establishment method, and system | |
CN111493850A (en) | Blood pressure measuring method and device based on deep neural network | |
US8197415B2 (en) | Non-invasive heart monitoring apparatus and method | |
Rajala et al. | Pulse arrival time (PAT) measurement based on arm ECG and finger PPG signals-comparison of PPG feature detection methods for PAT calculation | |
CN108478203A (en) | A kind of blood pressure measuring method monitoring radar based on single vital sign | |
CN108354597A (en) | A kind of rapid blood pressure computational methods based on the extraction of optimal wave | |
CN110236508A (en) | A kind of non-invasive blood pressure continuous monitoring method | |
CN110604556A (en) | Noninvasive continuous blood pressure measuring device and measuring method | |
CN110881967A (en) | Non-invasive multi-segment peripheral arterial vessel elastic function detection method and instrument thereof | |
CN112089405A (en) | Pulse wave characteristic parameter measuring and displaying device | |
CN108926334A (en) | Blood pressure acquisition methods and its system and device based on pulse wave | |
CN110840428B (en) | Noninvasive blood pressure estimation method based on one-dimensional U-Net network | |
CN113499059A (en) | BCG signal processing system and method based on optical fiber sensing non-contact | |
Tjahjadi et al. | Review of photoplethysmography based non-invasive continuous blood pressure methods | |
CN111839488A (en) | Non-invasive continuous blood pressure measuring device and method based on pulse wave | |
Zhang et al. | Development of a Continuous Blood Pressure Monitoring System based on Pulse Transit Time and Hemodynamic Covariates. | |
CN114041764A (en) | Method for estimating blood pressure without cuff based on brain impedance and electrocardiosignal | |
CN210095711U (en) | Noninvasive continuous blood pressure measuring equipment | |
Wang et al. | An improved algorithm for noninvasive blood pressure measurement |
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 |