CN109247929B - Blood pressure determination device, method, apparatus, and storage medium - Google Patents
Blood pressure determination device, method, apparatus, and storage medium Download PDFInfo
- Publication number
- CN109247929B CN109247929B CN201811421288.4A CN201811421288A CN109247929B CN 109247929 B CN109247929 B CN 109247929B CN 201811421288 A CN201811421288 A CN 201811421288A CN 109247929 B CN109247929 B CN 109247929B
- Authority
- CN
- China
- Prior art keywords
- pulse wave
- blood pressure
- pressure
- preset
- baseline
- 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.)
- Expired - Fee Related
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
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Cardiology (AREA)
- Veterinary Medicine (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Vascular Medicine (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The embodiment of the invention discloses a blood pressure determination device, a method, equipment and a storage medium, wherein the device comprises: the acquisition module is used for acquiring preset pulse wave characteristic points of a measured person; and the output module is used for inputting the preset pulse wave characteristic points into the trained blood pressure model to obtain and output the blood pressure of the person to be measured. The technical problem that the blood pressure measuring method in the prior art is not suitable for portable and wearable medical equipment is solved, and the technical effect that the portable blood pressure measurement has higher accuracy is achieved.
Description
Technical Field
The embodiment of the invention relates to the technical field of medical data processing, in particular to a blood pressure determination device, a blood pressure determination method, blood pressure determination equipment and a storage medium.
Background
Blood pressure is one of the important vital signs. The measurement of blood pressure is not only an important basis for clinical diagnosis and treatment of cardiovascular diseases, but also an important means for early prevention and early discovery of cardiovascular diseases in daily life. The blood pressure measuring method can be divided into invasive measurement, non-invasive measurement, intermittent measurement, continuous measurement and the like, wherein the continuous blood pressure measurement has important significance in the aspects of analyzing blood pressure variability, diagnosing potential hypertension and white coat hypertension, evaluating target organ damage, evaluating the curative effect of antihypertensive drugs and the like.
At present, the common noninvasive continuous blood pressure measuring methods include an arterial tension method, a volume clamp method and a pulse transit time method. Among them, the arterial tension method and the volume clamping method are complicated in equipment and complicated in operation, and are not suitable for portable and wearable medical equipment, nor for measurement outside hospitals. Since it is necessary to apply a certain pressure to the blood vessel, it causes a certain discomfort to the subject in long-term use, and is not suitable for long-term continuous measurement of blood pressure. The pulse transit time method overcomes the defects of the two methods, but has better effect only on measuring the systolic pressure and has larger measurement deviation on the diastolic pressure.
In summary, the blood pressure measurement methods of the prior art are not suitable for portable, wearable medical devices.
Disclosure of Invention
The embodiment of the invention provides a blood pressure determining device, a blood pressure determining method, blood pressure determining equipment and a storage medium, and aims to solve the technical problem that a measuring method in the prior art is not suitable for portable and wearable medical equipment.
In a first aspect, an embodiment of the present invention provides a blood pressure determining apparatus, including:
the acquisition module is used for acquiring preset pulse wave characteristic points of a measured person;
and the output module is used for inputting the preset pulse wave characteristic points into a trained blood pressure model so as to obtain and output the blood pressure of the tested person.
In a second aspect, an embodiment of the present invention further provides a blood pressure determining method, including:
acquiring preset pulse wave characteristic points of a measured person through an acquisition module;
and inputting the preset pulse wave characteristic points into a trained blood pressure model through an output module so as to obtain and output the blood pressure of the tested person.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the blood pressure determination method of the second aspect.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to perform the blood pressure determination method according to the second aspect.
The technical scheme of the blood pressure determining device provided by the embodiment of the invention comprises an acquisition module and an output module, wherein the acquisition module is used for acquiring the preset pulse wave characteristic points of a measured person; the output module is used for inputting the preset pulse wave characteristic points into the trained blood pressure model so as to obtain the blood pressure of the person to be measured. The blood pressure value is determined based on the high-precision pulse wave signals obtained by the portable equipment, so that the high-precision blood pressure value is obtained through an indirect method, and the portable measurement of the blood pressure is realized.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a block diagram of a blood pressure determination device according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a pulse wave according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a first pulse signal according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a second pulse signal according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a baseline provided by an embodiment of the present invention;
FIG. 6 is a schematic diagram of preset pulse feature points according to an embodiment of the present invention;
FIG. 7 is a schematic illustration of another baseline provided by an embodiment of the present invention;
FIG. 8 is a schematic diagram of another preset pulse feature point according to an embodiment of the present invention;
FIG. 9 is a flowchart of a blood pressure determination method according to a second embodiment of the present invention;
FIG. 10 is a flowchart of a blood pressure model training method according to a third embodiment of the present invention;
FIG. 11 is a schematic diagram showing the correlation between the estimated value and the measured value of systolic blood pressure according to the third embodiment of the present invention;
FIG. 12 is a schematic diagram illustrating the correlation between the estimated diastolic pressure and the measured diastolic pressure according to the third embodiment of the present invention;
fig. 13 is a schematic diagram illustrating the correlation between the estimated value and the measured value of the average pressure according to the third embodiment of the present invention;
fig. 14 is a block diagram of an apparatus according to a fourth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described through embodiments with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Fig. 1 is a block diagram of a blood pressure determination device according to a first embodiment of the present invention. The device comprises an acquisition module 11 and an output module 12, wherein the acquisition module 11 is used for acquiring preset pulse wave characteristic points of a measured person; the output module 12 is configured to input the preset pulse wave feature points into the trained blood pressure model to obtain and output the blood pressure of the subject.
The existing blood pressure monitoring methods usually directly measure blood pressure parameters, but the methods are not suitable for portable measurement. In order to adapt to portable blood pressure measurement, the blood pressure value of the measured person is indirectly determined through the pulse wave. Therefore, when measuring blood pressure, it is necessary to first acquire the pulse wave of the subject (see fig. 2), and then estimate the blood pressure value by the pulse wave.
Since the pulse wave is a continuous waveform and the data amount is large, in order to increase the speed of blood pressure measurement, the present embodiment uses the preset pulse feature points of the pulse wave to estimate the blood pressure. The determination method of the preset pulse wave feature points may be: the pulse wave is acquired by the feature point extraction unit 111 and the feature points of the acquired pulse wave are extracted, and the feature points are normalized by the normalization unit 112 to obtain preset pulse wave feature points.
Wherein, the extraction mode of the characteristic points of the pulse wave can be selected as follows: the position determining subunit determines the peak position and the trough position of the pulse wave, and the feature point extracting subunit extracts the feature points of the pulse wave according to the peak position and the trough position. Wherein, the position determining subunit respectively passes the pulse wave through a first band pass filter and a second band pass filter to respectively obtain a first pulse signal (see fig. 3) and a second pulse signal (see fig. 4), wherein the bandwidth of the first band pass filter is smaller than that of the second band pass filter, for example, the band pass range of the first band pass filter is 0.45-2Hz, and the band pass range of the second band pass filter is 0.45-20Hz, because the bandwidth of the first band pass filter is smaller than that of the second band pass filter, the first pulse signal is a coarse signal compared with the second pulse signal, the signal is a severely distorted pulse wave signal, the detailed information of the pulse wave such as the dicrotic wave, the dicrotic notch and the like is removed, and only the approximate waveform similar to a sine wave is kept, therefore, the position of the maximum value or the minimum value of the first pulse wave can be detected by adopting a difference method or a threshold segmentation method and the like, the estimated peak position or the estimated valley position of the pulse wave, then the rough central position detected in the second pulse signal is the estimated central position, then the maximum value or the minimum value in the preset neighborhood range of the estimated central position is obtained, and the peak position or the valley position of the pulse wave can be further obtained, wherein the neighborhood range of 0.25-0.4s can be selected as the preset neighborhood range, and if the preset neighborhood range is 0.3s, the pulse wave signal of 0.15s is obtained before and after by taking the estimated central position as the center.
The normalization of the feature points is to convert the amplitude of the feature points to a range of 0-1, and in order to improve the accuracy of the feature points, the influence of the baseline on the pulse wave is generally considered, for this reason, the obtaining module 11 further includes a first baseline determining unit 1131 and a second baseline determining unit 1132, and the two baseline determining units respectively correspond to the two baseline determining manners.
The first baseline determination unit 1131 is specifically configured to: the second pulse signal of the pulse wave is divided into a plurality of segments by taking the wave trough position as a dividing point, each segment corresponds to one pulse wave (namely one heartbeat), and each pulse wave comprises a wave crest point and two wave trough points. The height from the first valley point to the peak is defined as the rising height, and the height from the second valley point to the peak is defined as the falling height. A data point where each pulse wave increases the rise height by 10% from the first valley point along the time axis is defined as a first base point (start point), and a data point where each pulse wave increases the fall height by 10% from the second valley point along the time axis is defined as a second base point (end point). A straight line passing through the first base point and the second base point is defined as a base line (see the broken line in fig. 5 in particular). Corresponding to the first baseline determination unit 1131, the normalization unit 112 first intercepts the waveform signal between the starting point and the ending point, and then resamples each pulse wave to a preset number of data points, such as 50, 70, or 100 data points, by using cubic spline interpolation to remove the influence of the heart rate variation on the pulse wave length, so that all pulse waves have the same length; then subtracting the baseline to zero the first base point and the second base point; and then divided by the maximum value of the waveform at that time to set the amplitude of the peak to 1, so that the normalized pulse wave characteristic point, i.e., the preset pulse wave characteristic point, is obtained, as shown in fig. 6. .
It is understood that the normalization unit may also subtract the baseline from the pulse wave between the first base point and the second base point before normalizing the pulse wave feature points, then extract the feature points of each pulse wave, and then divide the feature points of the pulse wave by the maximum value of the waveform to obtain the preset pulse wave feature points. It can be understood that the peak amplitude of the preset pulse wave feature point is 1.
The second baseline determination unit 1132 is specifically configured to: the peak position is used as a dividing point to divide the fine signal of the pulse wave into a plurality of segments, and each segment corresponds to one pulse wave (namely one heartbeat). Each pulse wave includes two peak points and one valley point because the peaks are used as the division points, and the first peak point is used as the first base point and the second peak point is used as the second base point. A straight line connecting the first base point and the second base point is defined as a base line, see fig. 7. Corresponding to the second baseline determination unit 1132, the normalization unit 112 first intercepts the waveform signal between the first peak point and the second peak point, and resamples it to a preset number of data points, such as 50, 70, or 100 data points, by using cubic spline interpolation, so that all pulse waves have the same sampling point number; then subtracting the base line to make the first peak point and the second peak point set to zero; then, the pulse wave waveform is divided by the minimum value of the waveform (the minimum value is negative and the absolute value is maximum because the baseline is the connecting line of the wave crest and the waveform is negative after the baseline is subtracted), so that the pulse wave waveform is turned upwards, and the amplitude value at the minimum value of the original waveform is 1, so that the preset pulse wave characteristic point is obtained, as shown in fig. 8.
The wave crests are used as the dividing points, the wave troughs are positioned in the middle, the problem that the dividing is inaccurate due to the flat wave forms near the wave troughs can be avoided, and a better effect can be achieved. However, since the intercepted single pulse wave actually spans two pulse waves in tandem, including the falling part of the previous pulse wave and the rising part of the next pulse wave, it does not correspond to one cardiac cycle in the conventional sense, but also has higher blood pressure prediction accuracy.
It can be understood that before extracting the feature points of the pulse wave, the acquired pulse wave generally needs to be preprocessed, so the obtaining module of the embodiment further includes the preprocessing unit 114, but the embodiment does not limit the specific implementation form of the preprocessing unit, and the prior art is adopted.
Since the blood pressure parameters include at least diastolic pressure and systolic pressure, in order to improve the accuracy of each blood pressure parameter, the trained blood pressure model of the present embodiment includes a regression model that estimates each blood pressure parameter, for example, if the blood pressure parameters include diastolic pressure and systolic pressure, the trained blood pressure model includes a diastolic pressure regression model and a systolic pressure regression model. Correspondingly, the output module 12 inputs the preset pulse wave feature points into the trained blood pressure model, that is, the preset pulse wave feature points are respectively input into the trained diastolic pressure regression model and systolic pressure regression model to obtain the diastolic pressure and the systolic pressure of the measured person; if the blood pressure parameters are diastolic pressure, systolic pressure and average pressure, the trained blood pressure model includes a diastolic pressure regression model, a systolic pressure regression model and an average pressure regression model, and correspondingly, the output module 12 inputs the preset pulse wave feature points to the trained blood pressure model, that is, the preset pulse wave feature points are respectively input to the trained diastolic pressure regression model, systolic pressure regression model and average pressure regression model, so as to obtain the diastolic pressure, the systolic pressure and the average pressure of the person to be tested.
It can be understood that the trained blood pressure model can estimate the blood pressure through the pulse wave, and is usually based on the connection or relationship between the pulse wave and the blood pressure data, and to establish such connection or relationship, it is necessary to determine the blood pressure model, then obtain the pulse wave and the blood pressure data of the subject at the same time, and train the blood pressure model based on the pulse wave and the blood pressure data obtained at the same time, so the blood pressure determining apparatus of this embodiment further includes a blood pressure model determining module 13.
The blood pressure model determining module 13 is used for simultaneously obtaining the pulse wave and blood pressure data of the tested person; determining preset pulse wave characteristic points of the pulse waves and corresponding relations between the preset pulse wave characteristic points and the diastolic pressure, the systolic pressure and the average pressure of the blood pressure data; and establishing a diastolic pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the diastolic pressure, establishing a systolic pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the diastolic pressure, and establishing an average pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the average pressure on the basis of a support vector machine.
Wherein, the pulse wave of this embodiment is acquireed and is adopted prior art can, for example, utilize optical sensor to gather the photoplethysmography pulse wave signal of finger, perhaps utilize pressure sensor to gather the pressure pulse signal of wrist, and the sampling rate should be greater than 50 Hz. The blood pressure meter can use invasive blood pressure measuring equipment, and also can use equipment for continuously measuring blood pressure by using principles such as an arterial tension method, a volume clamp method or a pulse conduction time method, and the like, and at least a systolic pressure value and a diastolic pressure value can be ensured to be given to each heart beat. In order to improve the stability of the blood pressure model and the accuracy of blood pressure estimation, in the present embodiment, in the blood pressure and pulse wave collecting process, stimulation methods such as cold water, fist making movement or sound and image may be adopted to make the blood pressure have a certain fluctuation.
The corresponding relationship between the pulse wave and the blood pressure data can be expressed by that the preset pulse wave characteristic point is an independent variable, and the blood pressure data is a dependent variable. The blood pressure model usually includes at least a diastolic pressure regression model and a systolic pressure regression model established based on a support vector machine, and may further include a combination of mean pressure regression models. And the support vector machine preferably adopts an open source LIBSVM toolkit of professor Taiwan Chile, and the kernel function is preferably a radial basis kernel function.
After the blood pressure model is determined, when the blood pressure model is to be used to estimate the blood pressure according to the pulse wave, the blood pressure model also needs to be trained, and the blood pressure model includes a diastolic pressure regression model, a systolic pressure regression model and a mean pressure regression model as examples. The user or the manufacturer needs to determine the number of preset training samples, and input the preset pulse wave feature points and the blood pressure data of the preset sample number into the diastolic pressure regression model, the systolic pressure regression model and the average pressure regression model for model training to generate a trained diastolic pressure regression model, a trained systolic pressure regression model and a trained average pressure regression model.
In summary, according to the technical scheme of the blood pressure determining method provided by the embodiment of the invention, the preset pulse wave feature points of the measured person are obtained through the obtaining module; and inputting the preset pulse wave characteristic points into the trained blood pressure model through an output module so as to obtain the blood pressure of the tested person. The blood pressure value is determined based on the high-precision pulse wave signals obtained by the portable equipment, so that the high-precision blood pressure value is obtained through an indirect method, and the portable measurement of the blood pressure is realized.
Example two
Fig. 9 is a flowchart of a blood pressure determining method according to a second embodiment of the present invention. The technical scheme of the embodiment is suitable for the portable blood pressure measurement of the detected person. The method can be executed by the blood pressure determination device provided by the embodiment of the invention, and the device can be realized in a software and/or hardware manner and is configured to be applied in a processor. The method specifically comprises the following steps:
s101, acquiring preset pulse wave characteristic points of the measured person through an acquisition module.
The existing blood pressure monitoring methods usually directly measure blood pressure parameters, but the methods are not suitable for portable measurement. In order to adapt to portable blood pressure measurement, the blood pressure value of the measured person is indirectly determined through the pulse wave. Therefore, when measuring blood pressure, it is necessary to first acquire the pulse wave of the subject (see fig. 2), and then estimate the blood pressure value by the pulse wave.
Since the pulse wave is a continuous waveform and the data amount is large, in order to increase the blood pressure measurement speed, the present embodiment uses the preset pulse feature points acquired by the acquisition module 11 to perform blood pressure estimation. The method for determining the preset pulse wave characteristic points comprises the following steps: the pulse wave is acquired by the feature point extraction unit 111 and the feature points of the acquired pulse wave are extracted, and the feature points are normalized by the normalization unit 112 to obtain preset pulse wave feature points.
The method for extracting the characteristic points of the pulse waves comprises the following steps: the position determining subunit determines the peak position and the trough position of the pulse wave, and the feature point extracting subunit extracts the feature points of the pulse wave according to the peak position and the trough position. The method for determining the positions of the wave crests and the wave troughs comprises the following steps: the pulse wave is respectively passed through a first band-pass filter and a second band-pass filter by a position determining subunit to respectively obtain a first pulse signal (see fig. 3) and a second pulse signal (see fig. 4), wherein the bandwidth of the first band-pass filter is smaller than that of the second band-pass filter, for example, the band-pass range of the first band-pass filter is 0.45-2Hz, the band-pass range of the second band-pass filter is 0.45-20Hz, and the first pulse signal is a coarse signal compared with the second pulse signal because the bandwidth of the first band-pass filter is smaller than that of the second band-pass filter, the signal is a severely distorted pulse wave signal, the detailed information such as the dicrotic wave, the dicrotic notch and the like of the pulse wave is removed, and only the approximate waveform like a sine wave is kept, so that the position of the maximum value or the minimum value of the first pulse wave can be detected by a difference method or a threshold segmentation method and the like, the estimated peak position or the estimated valley position of the pulse wave, then the rough central position detected in the second pulse signal is the estimated central position, then the maximum value or the minimum value in the preset neighborhood range of the estimated central position is obtained, and the peak position or the valley position of the pulse wave can be further obtained, wherein the neighborhood range of 0.25-0.4s can be selected as the preset neighborhood range, and if the preset neighborhood range is 0.3s, the pulse wave signal of 0.15s is obtained before and after by taking the estimated central position as the center.
The normalization of the feature points is to convert the amplitude of the feature points to a range of 0-1, and in order to improve the accuracy of the feature points, it is usually necessary to consider the influence of the baseline on the pulse wave, for this reason, the obtaining module 11 further includes a first baseline determining unit 1131 and a second baseline determining unit 1132, and the two baseline determining units respectively correspond to different baseline determining manners.
The first baseline determination unit 1131 is specifically configured to: the second pulse signal of the pulse wave is divided into a plurality of segments by taking the wave trough position as a dividing point, each segment corresponds to one pulse wave (namely one heartbeat), and each pulse wave comprises a wave crest point and two wave trough points. The height from the first valley point to the peak is defined as the rising height, and the height from the second valley point to the peak is defined as the falling height. A data point where each pulse wave increases the rise height by 10% from the first valley point along the time axis is defined as a first base point (start point), and a data point where each pulse wave increases the fall height by 10% from the second valley point along the time axis is defined as a second base point (end point). A straight line passing through the first base point and the second base point is defined as a base line (see the broken line in fig. 5 in particular). Corresponding to the first baseline determination unit 1131, the normalization unit 112 first intercepts the waveform signal between the starting point and the ending point, and then resamples each pulse wave to a preset number of data points, such as 50, 70, or 100 data points, by using cubic spline interpolation to remove the influence of the heart rate variation on the pulse wave length, so that all pulse waves have the same length; then subtracting the baseline to zero the first base point and the second base point; and then divided by the maximum value of the waveform at that time to set the amplitude of the peak to 1, thereby obtaining the normalized pulse wave characteristic point, i.e., the preset pulse wave characteristic point.
As shown in fig. 6. It is understood that the normalization unit may also subtract the baseline from the pulse wave between the first base point and the second base point before normalizing the pulse wave feature points, then extract the feature points of each pulse wave, and then divide the feature points of the pulse wave by the maximum value of the waveform to obtain the preset pulse wave feature points. It can be understood that the peak amplitude of the preset pulse wave feature point is 1.
The second baseline determination unit 1132 is specifically configured to: the peak position is used as a dividing point to divide the fine signal of the pulse wave into a plurality of segments, and each segment corresponds to one pulse wave (namely one heartbeat). Each pulse wave includes two peak points and one valley point because the peaks are used as the division points, and the first peak point is used as the first base point and the second peak point is used as the second base point. A straight line connecting the first base point and the second base point is defined as a base line, see fig. 7. Corresponding to the second baseline determination unit 1132, the normalization unit 112 first intercepts the waveform signal between the first peak point and the second peak point, and resamples it to a preset number of data points, such as 50, 70, or 100 data points, by using cubic spline interpolation, so that all pulse waves have the same sampling point number; then subtracting the base line to make the first peak point and the second peak point set to zero; then, the pulse wave waveform is divided by the minimum value of the waveform (the minimum value is negative and the absolute value is maximum because the baseline is the connecting line of the wave crest and the waveform is negative after the baseline is subtracted), so that the pulse wave waveform is turned upwards, and the amplitude value at the minimum value of the original waveform is 1, so that the preset pulse wave characteristic point is obtained, as shown in fig. 8.
The wave crests are used as the dividing points, the wave troughs are positioned in the middle, the problem that the dividing is inaccurate due to the flat wave forms near the wave troughs can be avoided, and a better effect can be achieved. However, since the intercepted single pulse wave actually spans two pulse waves in tandem, including the falling part of the previous pulse wave and the rising part of the next pulse wave, it does not correspond to one cardiac cycle in the conventional sense, but also has higher blood pressure prediction accuracy.
It is understood that, before feature point extraction is performed on the pulse wave, preprocessing is generally required to be performed on the acquired pulse wave, and the specific method of preprocessing is not limited in this embodiment.
And S102, inputting the preset pulse wave characteristic points into the trained blood pressure model through the output module to obtain and output the blood pressure of the tested person.
The blood pressure parameters at least include diastolic pressure and systolic pressure, and in order to improve the accuracy of each blood pressure parameter, the trained blood pressure model of the present embodiment includes a regression model estimating each blood pressure parameter, for example, if the blood pressure parameters include diastolic pressure and systolic pressure, the trained blood pressure model includes a diastolic pressure regression model and a systolic pressure regression model. Correspondingly, the output module 12 inputs the preset pulse wave feature points into the trained blood pressure model, that is, the preset pulse wave feature points are respectively input into the trained diastolic pressure regression model and systolic pressure regression model to obtain the diastolic pressure and the systolic pressure of the measured person; if the blood pressure parameters are diastolic pressure, systolic pressure and average pressure, the trained blood pressure model includes a diastolic pressure regression model, a systolic pressure regression model and an average pressure regression model, and correspondingly, the output module 12 inputs the preset pulse wave feature points to the trained blood pressure model, that is, the preset pulse wave feature points are respectively input to the trained diastolic pressure regression model, systolic pressure regression model and average pressure regression model, so as to obtain the diastolic pressure, the systolic pressure and the average pressure of the person to be tested.
The technical scheme of the blood pressure determination method provided by the embodiment of the invention comprises the following steps: acquiring preset pulse wave characteristic points of a measured person; and inputting the preset pulse wave characteristic points into the trained blood pressure model to obtain the blood pressure of the tested person. The blood pressure value is determined based on the high-precision pulse wave signals obtained by the portable equipment, so that the high-precision blood pressure value is obtained through an indirect method, and the portable measurement of the blood pressure is realized.
EXAMPLE III
Fig. 10 is a flowchart of a blood pressure model training method according to a third embodiment of the present invention. On the basis of the embodiment, the embodiment of the invention adds the steps of a blood pressure model training method, and the method comprises the following steps:
and S1001, simultaneously acquiring pulse wave and blood pressure data of the tested person.
The trained blood pressure model can estimate the blood pressure through the pulse wave, and is usually based on the relationship between the pulse wave and the blood pressure data, and if the relationship is established, the pulse wave and the blood pressure data of the tested person need to be acquired simultaneously.
Wherein, the pulse wave of this embodiment is acquireed and is adopted prior art can, for example, utilize optical sensor to gather the photoplethysmography pulse wave signal of finger, perhaps utilize pressure sensor to gather the pressure pulse signal of wrist, and the sampling rate should be greater than 50 Hz. The blood pressure meter can use invasive blood pressure measuring equipment, and also can use equipment for continuously measuring blood pressure by using principles such as an arterial tension method, a volume clamp method or a pulse conduction time method, and the like, and at least a systolic pressure value and a diastolic pressure value can be ensured to be given to each heart beat. In order to improve the stability of the blood pressure model and the accuracy of blood pressure estimation, in the present embodiment, in the blood pressure and pulse wave collecting process, stimulation methods such as cold water, fist making movement or sound and image may be adopted to make the blood pressure have a certain fluctuation.
S1002, determining preset pulse wave characteristic points of pulse waves and corresponding relations between the preset pulse wave characteristic points and diastolic pressure, systolic pressure and average pressure of blood pressure data.
And establishing a corresponding relation between the preset pulse wave characteristic points of the tested person and the blood pressure data by taking the preset pulse wave characteristic points as independent variables and taking the blood pressure data as dependent variables.
S1003, based on a support vector machine, establishing a diastolic pressure regression model according to the corresponding relation between the preset pulse wave feature point and the diastolic pressure, establishing a systolic pressure regression model according to the corresponding relation between the preset pulse wave feature point and the diastolic pressure, and establishing an average pressure regression model according to the corresponding relation between the preset pulse wave feature point and the average pressure.
The blood pressure model usually includes at least a diastolic pressure regression model and a systolic pressure regression model established based on a support vector machine, and may further include a combination of mean pressure regression models. The blood pressure model including the diastolic pressure regression model, the systolic pressure regression model and the average pressure regression model is taken as an example for explanation. Firstly, the number of preset training samples is determined, and the preset pulse wave feature points and blood pressure data of the preset samples are combined and input into a diastolic pressure regression model, a systolic pressure regression model and a mean pressure regression model for model training so as to generate a trained diastolic pressure regression model, a trained systolic pressure regression model and a trained mean pressure regression model. The support vector machine preferably adopts an open source LIBSVM tool package of professor Taiwan Chile, and the kernel function is preferably a radial basis kernel function.
Illustratively, a Medical pulse oximeter is used for collecting pulse waves of 70 healthy subjects, and simultaneously, continuous blood pressure measuring instruments Finapres (Finapres Medical Systems b.v., the netherlands) are used for measuring systolic pressure, diastolic pressure and average pressure of each heart beat of the subject, and the cold water stimulation is used for generating certain fluctuation of the blood pressure in the experiment.
The accuracy of the regression model was verified using a 10-fold cross-validation method. Firstly, dividing the pulse wave and blood pressure data of each tested person into 10 sub-sample sets with equal size (the input vector is a preset pulse wave characteristic point, and the target values are systolic pressure, diastolic pressure and average pressure), wherein 9 sub-sets are used for training a support vector machine regression model, and a single sub-set is left as the precision of a test data verification model. Then, another 9 subsets were selected for training the blood pressure model, leaving a single subset as test data to verify the accuracy of the blood pressure model. And so on, repeating for a total of 10 times, each subset is verified once,
the present embodiment exemplarily presents the verification results of one of the verifications, such as fig. 11, such as 12, and fig. 13, wherein fig. 11 shows the relationship between the estimated value of diastolic pressure and the measured value, fig. 12 shows the relationship between the estimated value of systolic pressure and the measured value, and fig. 13 shows the relationship between the estimated value of average pressure and the measured value. In order to better describe the blood pressure estimation precision of the trained blood pressure model, statistics such as a Correlation Coefficient (CC), a mean square error (RMSE), a Mean Error (ME) and a Standard Deviation (SD) of an error of the blood pressure of each measured person are calculated, and the calculation formula is as follows:
wherein y represents an estimated value of the blood pressure model, and specifically is as follows: if the blood pressure model is a diastolic pressure regression model, y represents a diastolic pressure estimated value, if the blood pressure model is a systolic pressure regression model, y represents a systolic pressure estimated value, and if the blood pressure model is an average pressure regression model, y represents an average pressure estimated value; x represents the measured value of the continuous blood pressure measuring instrument, and specifically comprises the following components: if the blood pressure model is a diastolic pressure regression model, x represents the measured value of diastolic pressure, if the blood pressure model is a systolic pressure regression model, x represents the measured value of systolic pressure, and if the blood pressure model is an average pressure regression model, x represents the measured value of average pressure; n represents the sample volume, i.e. the number of valid heartbeats per subject that can be used for calculation.
The calculation results are summarized in table one, specifically as follows:
comparison of blood pressure estimated by the blood pressure model with measured blood pressure
Wherein SBP is systolic pressure, DBP is diastolic pressure, and MBP is mean pressure.
Table one shows that the blood pressure data obtained by using the trained blood pressure model of the present embodiment has higher accuracy, and therefore, it is expected that the blood pressure data has higher practicability in the field of portable blood pressure measurement.
Example four
Fig. 14 is a schematic structural diagram of an apparatus according to a fourth embodiment of the present invention, as shown in fig. 14, the apparatus includes a processor 201, a memory 202, an input device 203, and an output device 204; the number of the processors 201 in the device may be one or more, and one processor 201 is taken as an example in fig. 14; the processor 201, the memory 202, the input device 203 and the output device 204 in the apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 14.
The memory 202, as a computer-readable storage medium, may be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules (e.g., the acquisition module 11 and the output module 12) corresponding to the blood pressure determination method in the embodiment of the present invention. The processor 201 executes various functional applications of the device and data processing by running software programs, instructions and modules stored in the memory 202, i.e. implements the blood pressure determination method described above.
The memory 202 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 202 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 202 may further include memory located remotely from the processor 201, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 203 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the apparatus.
The output device 204 may include a display device such as a display screen, for example, of a user terminal.
EXAMPLE five
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for blood pressure determination, the method including:
acquiring preset pulse wave characteristic points of a measured person through an acquisition module;
and inputting the preset pulse wave characteristic points into a trained blood pressure model through an output module so as to obtain and output the blood pressure of the tested person.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the blood pressure determination method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a device (which may be a personal computer, a server, or a network device) to execute the blood pressure determining method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the blood pressure determining apparatus, the units and modules included in the embodiment are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (4)
1. A blood pressure determination device, comprising:
the acquisition module is used for acquiring preset pulse wave characteristic points of a measured person;
the output module is used for inputting the preset pulse wave characteristic points into a trained blood pressure model so as to obtain and output the blood pressure of the measured person;
the acquisition module includes:
the characteristic point extraction unit is used for acquiring the pulse wave of the tested person and extracting the characteristic points of the pulse wave;
the normalization unit is used for normalizing the characteristic points to obtain preset pulse wave characteristic points;
the feature point extraction unit includes:
the position determining subunit is used for acquiring the pulse wave of the tested person and determining the peak position and the trough position of the pulse wave;
a feature point extracting subunit, configured to extract feature points of the pulse wave according to the peak position and the trough position;
the position determining subunit is specifically configured to obtain a pulse wave of the subject, and pass the pulse wave through a first band-pass filter and a second band-pass filter, so as to obtain a first pulse signal and a second pulse signal, respectively; determining an estimated peak position and an estimated valley position of the pulse wave according to the first pulse signal; determining the estimated central point position of the pulse wave according to the second pulse wave signal; and determining the peak position and the trough position of the pulse wave according to the estimated central point position, the estimated peak position and the estimated trough position, wherein the bandwidth of the first band-pass filter is smaller than that of the second band-pass filter.
2. The apparatus of claim 1, wherein the obtaining module further comprises:
the first baseline determining unit is used for taking points which are respectively higher than two wave troughs by a first preset height and a second preset height as two base points when the pulse wave comprises the two wave troughs, and taking a connecting line of the two base points as a baseline;
accordingly, the normalization unit is specifically configured to: and taking the difference value of the characteristic point and the baseline as a characteristic point after baseline removal, and dividing the characteristic point after baseline removal by the maximum value of the current pulse wave to obtain a preset pulse wave characteristic point.
3. The apparatus of any of claims 1-2, wherein the obtaining module further comprises:
a second baseline determining subunit, configured to use, when the pulse wave includes two peaks, a connection line between the two peaks of the pulse wave as a baseline;
accordingly, the normalization unit is specifically configured to: and taking the difference value of the characteristic point and the baseline as a characteristic point after baseline removal, and dividing the characteristic point after baseline removal by the minimum value of the current pulse wave to obtain a preset pulse wave characteristic point.
4. The device of claim 3, further comprising a blood pressure model determination module for simultaneously acquiring pulse waves and blood pressure data of the subject; determining a preset pulse wave characteristic point of the pulse wave, and determining a corresponding relation between the preset pulse wave characteristic point and the diastolic pressure, the systolic pressure and the average pressure of the blood pressure data; and based on a support vector machine, establishing a diastolic pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the diastolic pressure, establishing a systolic pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the diastolic pressure, and establishing an average pressure regression model according to the corresponding relation between the preset pulse wave characteristic point and the average pressure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811421288.4A CN109247929B (en) | 2018-11-26 | 2018-11-26 | Blood pressure determination device, method, apparatus, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811421288.4A CN109247929B (en) | 2018-11-26 | 2018-11-26 | Blood pressure determination device, method, apparatus, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109247929A CN109247929A (en) | 2019-01-22 |
| CN109247929B true CN109247929B (en) | 2021-04-27 |
Family
ID=65042154
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811421288.4A Expired - Fee Related CN109247929B (en) | 2018-11-26 | 2018-11-26 | Blood pressure determination device, method, apparatus, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109247929B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021068193A1 (en) * | 2019-10-11 | 2021-04-15 | 中国科学院深圳先进技术研究院 | Method and apparatus for monitoring blood pressure waveform |
| CN113229788B (en) * | 2021-03-26 | 2022-07-19 | 广东粤港澳大湾区黄埔材料研究院 | Pulse wave denoising method and device based on film pressure sensor |
| CN116942102B (en) * | 2023-09-19 | 2023-12-22 | 北京清雷科技有限公司 | Sleep stage method and device based on pulse wave |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103908236A (en) * | 2013-05-13 | 2014-07-09 | 天津点康科技有限公司 | Automatic blood pressure measuring system |
| CN105286815A (en) * | 2015-11-02 | 2016-02-03 | 重庆大学 | Pulse wave signal feature point detection method based on waveform time domain features |
| CN106037694A (en) * | 2016-05-13 | 2016-10-26 | 吉林大学 | Continuous blood pressure measuring device based on pulse waves |
| CN107530005A (en) * | 2015-02-09 | 2018-01-02 | 日东电工株式会社 | Method and apparatus for the mean arterial pressure of derived object |
| CN107669256A (en) * | 2017-11-13 | 2018-02-09 | 深圳先进技术研究院 | Measuring method, the device and system of continuous blood pressure |
| CN108261191A (en) * | 2016-12-30 | 2018-07-10 | 深圳先进技术研究院 | Continuous BP measurement method, apparatus and equipment |
| CN108403100A (en) * | 2018-02-09 | 2018-08-17 | 重庆东渝中能实业有限公司 | Blood pressure computational methods based on oscillation wave waveform envelope feature |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4384047B2 (en) * | 2002-10-30 | 2009-12-16 | ディピーコム エイエス | A method for the analysis of single pulse pressure waves. |
| US7651466B2 (en) * | 2005-04-13 | 2010-01-26 | Edwards Lifesciences Corporation | Pulse contour method and apparatus for continuous assessment of a cardiovascular parameter |
| JP5688309B2 (en) * | 2011-02-23 | 2015-03-25 | フクダ電子株式会社 | Biological information analysis apparatus, biological information analysis method, and biological information analysis program |
| AU2013201556B2 (en) * | 2012-07-13 | 2014-06-05 | Gambro Lundia Ab | Filtering of pressure signals for suppression of periodic pulses |
| JP6418131B2 (en) * | 2015-10-20 | 2018-11-07 | 株式会社デンソー | Pulse wave signal processor |
| JP6829990B2 (en) * | 2016-12-26 | 2021-02-17 | 日本光電工業株式会社 | Atrial fibrillation detector, atrial fibrillation detection method, program and storage medium |
| CN108652605A (en) * | 2018-03-27 | 2018-10-16 | 上海交通大学 | Real-time blood pressure monitoring device based on single channel PPG signals |
| CN108720821A (en) * | 2018-05-28 | 2018-11-02 | 京东方科技集团股份有限公司 | A kind of blood pressure estimation device and blood pressure estimating system |
-
2018
- 2018-11-26 CN CN201811421288.4A patent/CN109247929B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103908236A (en) * | 2013-05-13 | 2014-07-09 | 天津点康科技有限公司 | Automatic blood pressure measuring system |
| CN107530005A (en) * | 2015-02-09 | 2018-01-02 | 日东电工株式会社 | Method and apparatus for the mean arterial pressure of derived object |
| CN105286815A (en) * | 2015-11-02 | 2016-02-03 | 重庆大学 | Pulse wave signal feature point detection method based on waveform time domain features |
| CN106037694A (en) * | 2016-05-13 | 2016-10-26 | 吉林大学 | Continuous blood pressure measuring device based on pulse waves |
| CN108261191A (en) * | 2016-12-30 | 2018-07-10 | 深圳先进技术研究院 | Continuous BP measurement method, apparatus and equipment |
| CN107669256A (en) * | 2017-11-13 | 2018-02-09 | 深圳先进技术研究院 | Measuring method, the device and system of continuous blood pressure |
| CN108403100A (en) * | 2018-02-09 | 2018-08-17 | 重庆东渝中能实业有限公司 | Blood pressure computational methods based on oscillation wave waveform envelope feature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109247929A (en) | 2019-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107530005B (en) | Method and device for deriving mean arterial pressure of a subject | |
| CN109247929B (en) | Blood pressure determination device, method, apparatus, and storage medium | |
| US11006874B2 (en) | Real-time stress determination of an individual | |
| US11006885B2 (en) | Apparatus for determining blood pressure | |
| CN111797817B (en) | Emotion recognition method, emotion recognition device, computer equipment and computer readable storage medium | |
| CN110755059B (en) | Blood pressure waveform monitoring method and device | |
| Yang et al. | Estimation and validation of arterial blood pressure using photoplethysmogram morphology features in conjunction with pulse arrival time in large open databases | |
| US10765332B2 (en) | Detection of the heartbeat in cranial accelerometer data using independent component analysis | |
| EP3061391A1 (en) | Apparatus and method for determining blood pressure | |
| WO2018129822A1 (en) | Method for measuring blood pressure and device | |
| CN101990445A (en) | Blood pressure estimating device and blood pressure estimating method | |
| CN105105734A (en) | Noninvasive continuous blood pressure measurement method, device and system based on heart sound signals | |
| CN108261192B (en) | Continuous blood pressure estimation method, device and equipment | |
| Lin et al. | Wave-shape oscillatory model for biomedical time series with applications | |
| Mejía-Mejía et al. | Comparison of pulse rate variability and morphological features of photoplethysmograms in estimation of blood pressure | |
| CN114159038A (en) | Blood pressure measuring method, device, electronic equipment and readable storage medium | |
| CN108430310A (en) | Assess the method and apparatus for carrying out this method of the reliability of blood pressure measurement | |
| WO2019153579A1 (en) | Electrocardiographic signal-based airbag-free blood pressure measurement method and system | |
| US20100312125A1 (en) | System for Cardiac Pathology Detection and Characterization | |
| CN108261191B (en) | Continuous blood pressure estimation method, device and equipment | |
| CN114929101A (en) | System and method for physiological measurement based on optical data | |
| Banerjee et al. | Blood pressure estimation from ECG data using XGBoost and ANN for wearable devices | |
| CN115003215A (en) | System and method for pulse transit time measurement from optical data | |
| Aysin et al. | Orthonormal-basis partitioning and time-frequency representation of cardiac rhythm dynamics | |
| Zhuang et al. | Remote blood pressure measurement via spatiotemporal mapping of a short-time facial video |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210427 Termination date: 20211126 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |