CN113208569A - Pulse wave curve fitting method based on group algorithm - Google Patents
Pulse wave curve fitting method based on group algorithm Download PDFInfo
- Publication number
- CN113208569A CN113208569A CN202010074596.5A CN202010074596A CN113208569A CN 113208569 A CN113208569 A CN 113208569A CN 202010074596 A CN202010074596 A CN 202010074596A CN 113208569 A CN113208569 A CN 113208569A
- Authority
- CN
- China
- Prior art keywords
- pulse wave
- algorithm
- bee
- fitting
- curve fitting
- 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.)
- Granted
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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4854—Diagnosis based on concepts of traditional oriental medicine
-
- 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
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Alternative & Traditional Medicine (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Fuzzy Systems (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Cardiology (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention provides a pulse wave curve fitting method based on a group algorithm. According to the motion characteristics of the blood in the double-elastic cavity, the blood circulation is simulated by adopting an equivalent circuit, a double-elastic cavity hemodynamics model is established, and then a global optimal solution is obtained according to an ABC algorithm in a group algorithm. The pulse wave curve fitting method based on the group algorithm can quickly and accurately diagnose the type II diabetes, and avoids misdiagnosis of the diagnosis result caused by subjective judgment of people.
Description
Technical Field
The invention relates to a pulse wave curve fitting method, in particular to a pulse wave curve fitting method based on a group algorithm
Background
Type II diabetes is generally slow in onset and is often seen in middle-aged and elderly adults. Early symptoms in such patients are not obvious and macrovascular and microvascular complications can often occur before definitive diagnosis.
Pulse taking is an important diagnosis method in traditional Chinese medicine and is an empirical summary of long-term medical practice of ancient medical scientists in China. The pulse type can be used as an important indicator of certain diseases, such as type II diabetes. At present, traditional Chinese medicine or clinical workers mainly rely on the pulse wave of the radial artery to judge the health state of the cardiovascular system.
However, the traditional Chinese medicine pulse diagnosis requires a doctor to perform long-term professional training, and the diagnosis result varies according to subjective judgment of people, so that the development of pulse diagnosis is limited. In the last thirty years, the pulse diagnosis objective research has achieved many beneficial results, and provides a basis for pulse condition signal analysis.
The analysis method of pulse condition information develops with the development of mathematics, biomechanics and engineering, and is restricted by the detection method of pulse condition information. At present, no rapid and effective pulse condition analysis method is specially used for diagnosing type II diabetes.
Disclosure of Invention
In order to solve the problems, the invention adopts the following technical scheme:
the invention provides a pulse wave curve fitting method of a group algorithm, which is used for fitting pulse wave signals of a radial artery of a hand of a patient, and is characterized by comprising the following steps: data acquisition, namely acquiring pulse wave signals of radial arteries of a hand of a patient by adopting a portable pulse wave acquisition platform consisting of a wearable wrist strap, a force sensor, an inflation and deflation circuit, an acquisition control module and a fitting module; establishing a dynamic model, simulating blood circulation by adopting an equivalent circuit according to the motion characteristics of blood in the double elastic cavities, and establishing a blood dynamic model in the double elastic cavities; fitting a waveform, namely fitting the waveform of the pulse wave according to a group algorithm theory; wherein, the equivalent circuit adopts an inductor L to simulate the inertia of blood and a capacitor C1And C2Simulating the left and right chambers of the patient's heart, and the resistance R simulating the total peripheral arterial impedanceForce, parameter R, C1、C2And L constitutes a vector ai。
The pulse wave curve fitting method of the group algorithm provided by the invention can also have the following characteristics: the portable pulse wave acquisition platform further comprises an air bag, an air pipe, a signal amplifier, a filter and an analog-to-digital converter.
The pulse wave curve fitting method of the group algorithm provided by the invention can also have the characteristics that the data acquisition comprises the following specific steps: presetting inflation pressure and storing in the acquisition control module, adopting charge-discharge circuit to exert pressure to the wearable wrist strap of wearing at patient's wrist, in case wearable wrist strap is aerifyd and is reached and predetermine inflation pressure, then acquisition control module just controls and fills gassing circuit and stop immediately aerifing, adopts force transducer to gather pulse wave signal of radial artery department simultaneously to pulse wave signal transmission who will gather gives the fitting module.
The pulse wave curve fitting method of the population algorithm provided by the invention can also have the characteristics that the specific steps for establishing the dynamic model are as follows: calculating the state equation according to the equivalent circuit of the aorta coupling system of the left cavity so as to obtain the blood flow at the root of the aorta as QA0Establishing a double-temperature Kerr model to simulate the blood flow from the aorta to the radial artery and the pulse waveform transmission, and setting the input blood flow Q of the systemin=QA0And obtaining the state equation of the quaternary bielastic cavity model:
wherein q is the radial blood flow, P1And P2The blood pressure of the aortic root and radial artery, respectively, R is the resistance of the peripheral system, C1And C2Is the approximate total compliance of the aorta and the peripheral system, L is the inertia of the blood flow,andare respectively q and P1And P2Variation in time.
The pulse wave curve fitting method of the group algorithm provided by the invention can also have the characteristics that the specific steps of fitting the waveform are as follows: judging a cost function for obtaining fitting quality, fitting the acquired pulse wave waveform by adopting a least square method, wherein the root mean square f (a) of the cost functioni) And calculating to obtain the optimal solution in the combination of the cost functions by adopting an ABC algorithm in a group intelligent algorithm.
The pulse wave curve fitting method of the group algorithm provided by the invention can also have the characteristics that the process of fitting the acquired pulse wave waveform by adopting a least square method is as follows: and calculating the root mean square of the cost function by adopting a least square method, wherein the calculation formula is as follows:
wherein, XsimFor arrays of simulation results, XexpFor the arrays of experimental results, N is the number of arrays.
The pulse wave curve fitting method of the group algorithm provided by the invention can also have the characteristics that the specific steps of calculating and obtaining the optimal solution in the combination of the cost functions by adopting the ABC algorithm in the group intelligent algorithm are as follows:
step S1, setting parameters of the ABC algorithm, including the number D of bee colonies, the maximum circulation time T and the vector aiSetting the current cycle number to be 0 at the boundary of the middle parameter,
step S2, initializing the employment bee, initializing the initial honey source position a according to the initialization formulaiAnd calculating the fitness of the solution, wherein an initialization formula is as follows:
aij=amin,j+rand(0,1)(amax,j-amin,j)
where i ∈ {1, 2.,. SN }, SN is the number of employed bees, j ∈ {1, 2.,. D }, the number of bee colonies DI.e. vector aiAnd rand (0,1) represents a random number within (0,1), amax,jAnd amin,jRespectively representing the maximum and minimum values of the feasible solution,
the fitness formula is as follows:
step S3, the employed bee generates new honey source n according to the root mean square of the cost functioniAnd calculating an adaptive value, wherein the calculation formula of the new honey source is as follows:
nij=aij+φik(aij-akj)
where k is in the form of {1, 2.,. m }, the variable k is randomly created, k ≠ i, φikIs a random value between-1 and 1,
step S4, determining vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S5, the hiring bee selects honey sources according to the greedy strategy,
step S6, calculating the probability P of honey source being selectediThe calculation formula is as follows:
step S7, the observation bee selects the employed bee by adopting a roulette method, carries out neighborhood search according to the root mean square of the cost function, generates a new honey source and calculates an adaptive value,
step S8, determining vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S9, the observation bee selects the honey source by adopting a greedy strategy,
step S10, judging whether there is abandoned honey source, if so, converting the employed bee into observation bee, and searching new honey source in solution space according to the initialization formula,
step S11, adding the current circulation times and recording the optimal solution,
and step S12, judging whether the current cycle frequency is larger than the maximum cycle frequency, if not, entering step S3, and if so, outputting the currently recorded optimal solution.
Action and Effect of the invention
According to the pulse wave curve fitting method of the group algorithm, a portable pulse wave acquisition platform consisting of a wearable wrist strap, a force sensor, a charging and discharging circuit and an acquisition control module is adopted, and pulse wave signals of the radial artery of the hand of a patient within a certain time are acquired under the action of a target force, so that pulse signals with variable depths are obtained. A blood dynamics theoretical model of a double elastic cavity is established to simulate the generation of heart pulses and the waveforms of the heart pulses. And performing theoretical fitting on the acquired signals by using the established model, and obtaining a global optimal solution through a group algorithm theory. The method can quickly and accurately diagnose the type II diabetes mellitus in an algorithm fitting mode.
Drawings
FIG. 1 is a portable pulse wave acquisition platform according to an embodiment of the present invention;
FIG. 2 is a bi-elastic chamber hemodynamic model of an embodiment of the present invention, wherein (a) is an equivalent circuit of the bi-elastic chamber model, (b) is a graph of blood flow velocity at the aortic root versus time, and (c) is a bi-thermocker model;
FIG. 3 is a flow chart of the ABC algorithm of the present invention computing an optimal solution in the combination of cost functions;
fig. 4 is a typical pulse waveform collected by the high-precision sensor according to the embodiment of the present invention, in which (a) is a pulse waveform composed of 1000 sampling points of patient number 32, and (b) to (e) are waveform diagrams after normalization of monocycles of patients number 32, 01, 19, and 28, respectively;
FIG. 5 is a graph of patient number 28 of an embodiment of the present invention, where graph (a) is the theoretically fit curve and the experimentally measured curve, the solid line is the experimentally measured curve, the circle is the theoretically fit curve, graph (b) is the minimum root mean square obtained through different iterations during the ABC algorithm, and graph (c) is the positional offset of the experimentally measured curve and the theoretically fit curve.
Detailed Description
The following description of the embodiments of the present invention will be made with reference to the accompanying drawings.
< example >
Fig. 1 is a portable pulse wave acquisition platform according to an embodiment of the invention.
As shown in fig. 1, the portable pulse wave acquisition platform comprises a wearable wrist strap, a force sensor, an airbag, a trachea, a processing module and a fitting module. The processing module comprises a signal amplifier, a filter, an analog-to-digital converter and an air charging and discharging circuit.
The method for acquiring the pulse wave signals of the radial artery of the hand of the patient by adopting the portable pulse wave acquisition platform comprises the following steps:
the inflation pressure is preset and stored in the acquisition control module.
The charging and discharging circuit is adopted to apply pressure to the wearable wrist strap worn on the wrist of the patient.
Once the wearable wrist strap is inflated to reach the preset inflation pressure, the acquisition control module controls the inflation and deflation circuit to immediately stop inflating, and meanwhile, the force sensor is used for acquiring pulse wave signals at the radial artery and sending the acquired pulse wave signals to the fitting module.
Fig. 2 is a diagram of a bi-elastic-lumen hemodynamic model according to an embodiment of the present invention, in which (a) is an equivalent circuit of the bi-elastic-lumen model, (b) is a diagram of a relationship between a blood flow velocity at an aortic root and time, and (c) is a bi-winker model.
As shown in fig. 2, a bielastic chamber hemodynamic model was established. As shown in FIG. 2 (a), the inertia of blood flow is represented by an inductance L disposed between two elastic chambers, and a capacitance C in the circuit is used1And C2Showing the left and right chambers. Due to the systolic and diastolic activity of the left ventricle, blood first enters the first lumen (i.e., the aorta) and the L-lead, then flows into the second lumen (i.e., the peripheral artery),finally, the vein is entered and the total peripheral arterial resistance is represented by resistance R. Aortic root flow QA0Calculated from the equation of state of the left ventricular aortic coupling system equivalent circuit.
As shown in fig. 2 (c), a two-temperature kerr model was established to simulate blood flow and pulse waveform transmission from the aorta to the radial artery. Input blood flow rate of the system Qin=QA0And obtaining the state equation of the quaternary bielastic cavity model:
wherein q is the radial blood flow, P1And P2The blood pressure of the aortic root and radial artery, respectively, R is the resistance of the peripheral system, C1And C2Is the approximate total compliance of the aorta and the peripheral system, L is the inertia of the blood flow,andare respectively q and P1And P2Variation in time.
And (3) theoretically fitting a pulse wave waveform by a group algorithm, and simulating various theoretical curves of the radial artery pulse waveform by adjusting parameters of the established double-elastic-cavity hemodynamics model.
The specific steps of fitting the waveform are as follows:
and judging a cost function of the fitting quality.
And (4) theoretical fitting, namely fitting by respectively adopting a least square method and a population algorithm.
Solving the root mean square f (a) of the cost function by using a least square methodi) Verifying the theoretical fitting effect of the simulation result and the experimental result, namely the root mean square f (a)i) The calculation formula is as follows:
wherein, aiIs given by the parameter R, C1、C2And a vector of L, XsimFor arrays of simulation results, XexpFor the arrays of experimental results, N is the number of arrays.
FIG. 3 is a flow chart of the ABC algorithm calculation of the embodiment of the present invention to obtain the optimal solution in the combination of cost functions.
As shown in fig. 3, the ABC algorithm in the group intelligence algorithm is used to calculate the optimal solution in the combination of the cost functions, and the specific steps are as follows:
step S1, setting parameters of the ABC algorithm, including the number D of bee colonies, the maximum circulation time T and the vector aiSetting the current cycle number to be 0 at the boundary of the middle parameter,
step S2, initializing the employment bee, initializing the initial honey source position a according to the initialization formulaiAnd calculating the fitness of the solution, wherein an initialization formula is as follows:
aij=amin,j+rand(0,1)(amax,j-amin,j)
in the formula, i is equal to {1, 2.,. SN }, SN is the number of employed bees, j is equal to {1, 2.,. D }, and the number D of bee groups is the vector aiAnd rand (0,1) represents a random number within (0,1), amax,jAnd amin,jRespectively representing the maximum and minimum values of the feasible solution,
the fitness formula is as follows:
step S3, the employed bee generates new honey source n according to the root mean square of the cost functioniAnd calculating an adaptive value, wherein the calculation formula of the new honey source is as follows:
nij=aij+φik(aij-akj)
where k is in the form of {1, 2.,. m }, the variable k is randomly created, k ≠ i, φikIs a random value between-1 and 1,
step S4Judging the vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S5, the hiring bee selects honey sources according to the greedy strategy,
step S6, calculating the probability P of honey source being selectediThe calculation formula is as follows:
step S7, the observation bee selects the employed bee by adopting a roulette method, carries out neighborhood search according to the root mean square of the cost function, generates a new honey source and calculates an adaptive value,
step S8, determining vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S9, the observation bee selects the honey source by adopting a greedy strategy,
step S10, judging whether there is abandoned honey source, if so, converting the employed bee into observation bee, and searching new honey source in solution space according to the initialization formula,
step S11, adding the current circulation times and recording the optimal solution,
and step S12, judging whether the current cycle frequency is larger than the maximum cycle frequency, if not, entering step S3, and if so, outputting the currently recorded optimal solution.
30 cases of type II diabetes and 52 healthy people are sampled by adopting a portable pulse wave acquisition platform, the number of sampling points is 1000, and the duration time is 10 s.
Fig. 4 is a typical pulse waveform acquired by the high-precision sensor according to the embodiment of the present invention, in which (a) is a pulse waveform composed of 1000 sampling points of patient number 32, and (b) to (e) are waveform diagrams after normalization of monocycles of patients number 32, 01, 19, and 28, respectively.
FIG. 5 is a graph of patient number 28 of an embodiment of the present invention, where graph (a) is the theoretically fit curve and the experimentally measured curve, the solid line is the experimentally measured curve, the circle is the theoretically fit curve, graph (b) is the minimum root mean square obtained through different iterations during the ABC algorithm, and graph (c) is the positional offset of the experimentally measured curve and the theoretically fit curve.
As shown in fig. 4, the root mean square value is the smallest when the positions of the experimentally measured curve and the theoretically fitted curve match.
And testing the root mean square cost time and the convergence rate of the established double-elastic cavity hemodynamic model by adopting a group algorithm. 20 and 30 iterations were selected for each trial, with the time cost and root mean square value shown in table 1. The cost time of 20 iterations is 284.83 + -0.79 s, and the minimum root mean square is 1.7324 + -0.0276. However, the cost time for 30 iterations is 419.44 ± 4.11s, and the minimum root mean square is 1.6901 ± 0.0016. Thus, the root mean square tends to be constant for 30 iterations.
TABLE 1
Furthermore, the root mean square of 1.6901 is approximately equal to the global optimal solution. And the ABC algorithm is adopted for theoretical fitting, and the optimal root mean square value of one period of the pulse waveform can be obtained within about 7 minutes.
A Support Vector Machine (SVM) method is adopted as a linear kernel classifier for supervised learning-based binary and multivariate classification. The bioelastic cavity hemodynamic model and the ABC algorithm are combined to obtain R, C1、C2And L the best of four parameters aiArray, which fits the experimental results with minimal cost. For the healthy and diabetic groups, the decision boundaries are found by using a single parameter or a combination of multiple parameters on the log group. The healthy and diabetic groups were first labeled as 0 and 1, respectively. To checkThe accuracy of the support vector machine method takes most of two groups as training items and the last 10 items as test items. The accuracy is defined by dividing the number of predicted and corrected tags by the total number of tags.
The results show that the blood flow inertia L and the total peripheral arterial compliance C of the diabetic patient are comparable to the established model2There are significant differences. The curve fitting method is not only effective, but also has a lower time cost. After 30 iterations, the classification results can be obtained in approximately 7 minutes. The result shows that the established model can be used as a simple and rapid method to realize the non-invasive prediction of early II type diabetes.
Examples effects and effects
The pulse wave curve fitting method based on the group algorithm provided by the embodiment adopts a portable pulse wave acquisition platform consisting of a wearable wrist strap, a force sensor, a charging and discharging circuit and an acquisition control module, and acquires pulse wave signals of radial arteries of a hand of a patient within a certain time under the action of a target force, so that pulse signals with variable depths are acquired. The generation of heart pulses and the waveforms of the heart pulses are simulated by establishing a double-elastic-cavity hemodynamics theoretical model, then the acquired signals are theoretically fitted by utilizing the established model, and a global optimal solution is obtained by a group algorithm theory, so that the type II diabetes can be rapidly and accurately diagnosed.
The above-described embodiments are merely illustrative of specific embodiments of the present invention, and the present invention is not limited to the description of the above-described embodiments.
Claims (7)
1. A pulse wave curve fitting method based on a group algorithm is used for fitting collected pulse wave signals of a radial artery of a hand of a patient, and is characterized by comprising the following steps:
data acquisition, namely acquiring the pulse wave signals of the radial artery of the hand of the patient by adopting a portable pulse wave acquisition platform consisting of a wearable wrist strap, a force sensor, an inflation and deflation circuit, an acquisition control module and a fitting module;
establishing a dynamic model, simulating blood circulation by adopting an equivalent circuit according to the motion characteristics of blood in the double elastic cavities, and establishing a blood dynamic model in the double elastic cavities;
fitting a waveform, namely fitting the waveform of the pulse wave according to a group algorithm theory;
wherein, the equivalent circuit adopts an inductor L to simulate the inertia of blood and a capacitor C1And C2Simulating the left and right chambers of the patient's heart, resistance R simulating the total peripheral arterial resistance, parameter R, C1、C2And L constitutes a vector ai。
2. The population algorithm-based pulse wave curve fitting method of claim 1,
the portable pulse wave acquisition platform further comprises an air bag, an air pipe, a signal amplifier, a filter and an analog-to-digital converter.
3. The population algorithm-based pulse wave curve fitting method according to claim 1, wherein the data acquisition comprises the following specific steps:
the inflation pressure is preset and stored in the acquisition control module,
a charging and discharging circuit is adopted to apply pressure to the wearable wrist strap worn on the wrist of the patient,
once the wearable wrist strap is inflated to reach the preset inflation pressure, the acquisition control module controls the inflation and deflation circuit to immediately stop inflating, and meanwhile, the force sensor is adopted to acquire pulse wave signals at the radial artery and send the acquired pulse wave signals to the fitting module.
4. The population algorithm-based pulse wave curve fitting method according to claim 1, wherein the specific steps of establishing the kinetic model are as follows:
calculating the state equation of the left chamber according to the equivalent circuit of the aorta coupling system of the left chamber so as to obtain the blood flow at the root of the aorta as QA0,
Establishing a double-temperature Kerr model to simulate the blood flow from the aorta to the radial artery and the pulse waveform transmission, and setting the input blood flow Q of the systemin=QA0And obtaining the state equation of the quaternary bielastic cavity model:
wherein q is the radial blood flow, P1And P2The blood pressure of the aortic root and radial artery, respectively, R is the resistance of the peripheral system, C1And C2Is the approximate total compliance of the aorta and the peripheral system, L is the inertia of the blood flow,andare respectively q and P1And P2Variation in time.
5. The population algorithm-based pulse wave curve fitting method according to claim 1, wherein the specific steps of fitting the waveform are as follows:
judging a cost function for obtaining the fitting quality,
fitting the acquired pulse wave waveform by adopting a least square method to obtain the root mean square f (a) of the cost functioni),
And calculating to obtain the optimal solution in the combination of the cost functions by adopting an ABC algorithm in a group intelligent algorithm.
6. The population algorithm-based pulse wave curve fitting method according to claim 5, wherein the least square method is adopted to fit the acquired pulse wave waveforms by:
and calculating the root mean square of the cost function by adopting the least square method, wherein the calculation formula is as follows:
wherein, XsimFor arrays of simulation results, XexpFor the arrays of experimental results, N is the number of arrays.
7. The population algorithm-based pulse wave curve fitting method according to claim 5, wherein the specific steps of calculating the optimal solution in the combination of the cost functions by using the ABC algorithm in the population intelligent algorithm are as follows:
step S1, setting parameters of the ABC algorithm, including the number D of bee colonies, the maximum circulation time T and the vector aiSetting the current cycle number to be 0 at the boundary of the middle parameter,
step S2, initializing the employment bee, initializing the initial honey source position a according to the initialization formulaiAnd calculating the fitness of the solution, wherein the initialization formula is as follows:
aij=amin,j+rand(0,1)(amax,j-amin,j)
wherein i is an element of {1, 2.,. SN }, SN is the number of the employed bees, j is an element of {1, 2.,. D }, and the number D of the bee colony is a vector aiAnd rand (0,1) represents a random number within (0,1), amax,jAnd amin,jRespectively representing the maximum and minimum values of the feasible solution,
the fitness formula is as follows:
step S3, the hiring bee generates new honey source n according to the root mean square of the cost functioniAnd calculating an adaptive value, wherein the calculation formula of the new honey source is as follows:
nij=aij+φik(aij-akj)
in the formulaK ∈ {1, 2., m }, the variable k is randomly created, k ≠ i, φ ≠ iikIs a random value between-1 and 1,
step S4, determining vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S5, the hiring bee selects honey sources according to a greedy strategy,
step S6, calculating the probability P of the honey source being selectediThe calculation formula is as follows:
step S7, the observation bee selects the employed bee by adopting a roulette method, carries out neighborhood search according to the root mean square of the cost function, generates a new honey source and calculates an adaptive value,
step S8, determining vector aiParameter R, C in (1)1、C2And whether L is out of bounds, if the L is judged to be out of bounds, setting the out-of-bounds parameter in the boundary range,
step S9, the observers adopt the greedy strategy to select the honey sources,
step S10, judging whether there is the abandoned honey source, if so, converting the employed bee into the observation bee, and searching a new honey source in a solution space at random according to the initialization formula,
step S11, adding the current circulation times and recording the optimal solution,
and step S12, judging whether the current cycle frequency is larger than the maximum cycle frequency, if not, entering step S3, and if so, outputting the currently recorded optimal solution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010074596.5A CN113208569B (en) | 2020-01-21 | 2020-01-21 | Pulse wave curve fitting method based on group algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010074596.5A CN113208569B (en) | 2020-01-21 | 2020-01-21 | Pulse wave curve fitting method based on group algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113208569A true CN113208569A (en) | 2021-08-06 |
CN113208569B CN113208569B (en) | 2022-06-10 |
Family
ID=77085414
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010074596.5A Active CN113208569B (en) | 2020-01-21 | 2020-01-21 | Pulse wave curve fitting method based on group algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113208569B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114010167A (en) * | 2021-11-25 | 2022-02-08 | 中山大学 | Pulse wave fitting method based on Weibull function |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000005139A (en) * | 1998-06-23 | 2000-01-11 | Seiko Epson Corp | Pulse wave detecting device and touch sense detecting device |
CN102917373A (en) * | 2012-10-23 | 2013-02-06 | 华南理工大学 | Method for calibrating and fitting indoor wireless transmission model based on artificial bee colony algorithm |
CN107045717A (en) * | 2017-03-21 | 2017-08-15 | 南京邮电大学 | The detection method of leucocyte based on artificial bee colony algorithm |
CN109820498A (en) * | 2019-01-28 | 2019-05-31 | 重庆邮电大学 | A kind of Wrist blood pressure meter that binding identity information and its pulse wave approximating method |
-
2020
- 2020-01-21 CN CN202010074596.5A patent/CN113208569B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000005139A (en) * | 1998-06-23 | 2000-01-11 | Seiko Epson Corp | Pulse wave detecting device and touch sense detecting device |
CN102917373A (en) * | 2012-10-23 | 2013-02-06 | 华南理工大学 | Method for calibrating and fitting indoor wireless transmission model based on artificial bee colony algorithm |
CN107045717A (en) * | 2017-03-21 | 2017-08-15 | 南京邮电大学 | The detection method of leucocyte based on artificial bee colony algorithm |
CN109820498A (en) * | 2019-01-28 | 2019-05-31 | 重庆邮电大学 | A kind of Wrist blood pressure meter that binding identity information and its pulse wave approximating method |
Non-Patent Citations (2)
Title |
---|
凌建东: "一种基于脉搏波的人体健康监护系统研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
张玮: "基于左心动脉和双弹性腔的脉象仿真研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114010167A (en) * | 2021-11-25 | 2022-02-08 | 中山大学 | Pulse wave fitting method based on Weibull function |
Also Published As
Publication number | Publication date |
---|---|
CN113208569B (en) | 2022-06-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gaurav et al. | Cuff-less PPG based continuous blood pressure monitoring—A smartphone based approach | |
Leitner et al. | Personalized blood pressure estimation using photoplethysmography: A transfer learning approach | |
CN102397064B (en) | Continuous blood pressure measuring device | |
CN102488503B (en) | Continuous blood pressure measurer | |
CN102429649B (en) | Continuous blood pressure measuring device | |
Tjahjadi et al. | Noninvasive classification of blood pressure based on photoplethysmography signals using bidirectional long short-term memory and time-frequency analysis | |
CN108577830B (en) | User-oriented physical sign information dynamic monitoring method and system | |
CN109044302A (en) | Measure device, electronic equipment and the computer readable storage medium of blood pressure | |
Maqsood et al. | A benchmark study of machine learning for analysis of signal feature extraction techniques for blood pressure estimation using photoplethysmography (PPG) | |
JP2016513516A (en) | Systems and methods for individual hemodynamic modeling and monitoring | |
CN110897633A (en) | Electrocardiosignal segmentation method based on deep neural network | |
US10729337B2 (en) | Device and method for non-invasive left ventricular end diastolic pressure (LVEDP) measurement | |
US20230218178A1 (en) | Construction method and application of digital human cardiovascular system based on hemodynamics | |
WO2021164345A1 (en) | Blood pressure prediction method and device | |
CN109497991A (en) | Heart failure detection device based on machine learning | |
CN115916056A (en) | Electrocardiogram generation device and method based on generation type antagonistic neural network algorithm | |
Paviglianiti et al. | Noninvasive arterial blood pressure estimation using ABPNet and VITAL-ECG | |
CN109599178A (en) | Threshold determination model determines method, apparatus, medical treatment detection device and storage medium | |
CN113208569B (en) | Pulse wave curve fitting method based on group algorithm | |
CN113907727B (en) | Beat-by-beat blood pressure measurement system and method based on photoplethysmography | |
CN113040738B (en) | Blood pressure detecting device | |
Shukla | Estimation of blood pressure from non-invasive data | |
US20220313167A1 (en) | Recliner having healthcare function | |
CN116138755A (en) | Method for constructing model for noninvasive blood pressure monitoring and wearable device | |
Shokouhmand et al. | Fingertip Strain Plethysmography: Representation of Pulse Information based on Vascular Vibration |
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 |