CN116650827A - Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network - Google Patents
Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network Download PDFInfo
- Publication number
- CN116650827A CN116650827A CN202310742773.6A CN202310742773A CN116650827A CN 116650827 A CN116650827 A CN 116650827A CN 202310742773 A CN202310742773 A CN 202310742773A CN 116650827 A CN116650827 A CN 116650827A
- Authority
- CN
- China
- Prior art keywords
- blood pump
- blood
- centrifugal
- neural network
- centrifugal blood
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M60/00—Blood pumps; Devices for mechanical circulatory actuation; Balloon pumps for circulatory assistance
- A61M60/30—Medical purposes thereof other than the enhancement of the cardiac output
- A61M60/36—Medical purposes thereof other than the enhancement of the cardiac output for specific blood treatment; for specific therapy
- A61M60/38—Blood oxygenation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M60/00—Blood pumps; Devices for mechanical circulatory actuation; Balloon pumps for circulatory assistance
- A61M60/20—Type thereof
- A61M60/205—Non-positive displacement blood pumps
- A61M60/216—Non-positive displacement blood pumps including a rotating member acting on the blood, e.g. impeller
- A61M60/226—Non-positive displacement blood pumps including a rotating member acting on the blood, e.g. impeller the blood flow through the rotating member having mainly radial components
- A61M60/232—Centrifugal pumps
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M60/00—Blood pumps; Devices for mechanical circulatory actuation; Balloon pumps for circulatory assistance
- A61M60/50—Details relating to control
- A61M60/508—Electronic control means, e.g. for feedback regulation
- A61M60/515—Regulation using real-time patient data
- A61M60/531—Regulation using real-time patient data using blood pressure data, e.g. from blood pressure sensors
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- Cardiology (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Anesthesiology (AREA)
- Mechanical Engineering (AREA)
- Hematology (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Epidemiology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Emergency Medicine (AREA)
- External Artificial Organs (AREA)
Abstract
An ECMO (electronic control unit) centrifugal blood pump flow pulsatility control system based on an RBF neural network realizes the pulsatility of the centrifugal blood pump blood flow by periodically changing the rotating speed of the centrifugal blood pump, and comprises a main controller module, a signal acquisition module, a motor driving module and a man-machine interaction module; the signal acquisition module is used for acquiring the inlet pressure and the outlet pressure of the current centrifugal blood pump and the intra-aortic blood pressure of the patient; the controller module calculates parameters such as tracking error, first-order differential of the tracking error and the like according to the actual aortic blood pressure of the patient and the expected aortic blood pressure, and calculates the rotating speed which the centrifugal blood pump should reach by using a preset RBF neural network self-adaptive control algorithm as the control quantity of the centrifugal blood pump-blood circulation coupling system according to the calculated parameters; the invention realizes the control of the flow pulsatility of the ECMO blood pump, and further realizes the accurate control of the aortic blood pressure waveform of the patient.
Description
Technical Field
The invention belongs to the technical field of self-adaptive control of complex nonlinear systems, and particularly relates to an ECMO (electronic control unit) centrifugal blood pump flow pulsatility control system based on an RBF (radial basis function) neural network.
Background
ECMO (Extracorporeal Membrane Oxygenation), which is commonly known as "She Kemo" and "artificial lung", is a medical emergency apparatus for providing sustained extracorporeal respiration and circulation to patients suffering from severe cardiopulmonary failure to sustain the life of the patient. When ECMO is operated, blood is drawn from veins, oxygenated through the membrane lung, and carbon dioxide is expelled, and oxygenated blood may be returned to veins (V-V transfer) or arteries (V-A transfer). The ECMO is an improved artificial heart-lung machine, the most core parts are a membrane lung and a blood pump, which respectively play the roles of the artificial lung and the artificial heart, can carry out long-time heart-lung support on patients suffering from severe heart-lung failure, and wins precious time for rescuing critical diseases.
The RBF neural network is a feedforward neural network, and the feedforward neural network is characterized in that the network under the structure has no feedback link, and the output of the former layer directly forms the input of the latter layer; the input layer of the network is directly formed by a source node, receives information input by the outside, and the output layer derives an output signal of the neural network to the outside; the input and output layers are directly connected with the outside, so that the input and output layers are called visible layers, and the structure between the input layer and the output layer is not directly connected with the outside, so that the input and output layers are called hidden layers.
The RBF neural network is named because the hidden layer activation function is a radial basis function, is a feedforward neural network with only one hidden layer, and has a three-layer overall network structure; the input layer is an interface between the front end of the neural network and the outside, and is responsible for receiving input signals and guiding the signals into the hidden layer; the hidden layer is a network core, and the hidden layer neurons perform nonlinear transformation on the received signals through an activation function, and send the transformed signals to the output layer through linear transformation; the output layer is an interface between the rear end of the neural network and the outside, and is responsible for integrating network output and exporting the network output to the outside.
The RBF neural network is characterized in that an input layer and an hidden layer are connected without weights, and an input signal is directly mapped to a hidden layer space through a radial basis function, so that mapping of the signal from low dimension to high dimension is realized, and the input signal is linearly separable in the high dimension; the mapping is nonlinear and the mapping relationship is only related to radial basis function parameters, and the adjustment of the input mapping can be realized by adjusting the radial basis function parameters. But the implicit layer and the output layer are connected by means of weights, i.e. the implicit layer space to output space mapping is linear. The mapping from input to output of the whole neural network is nonlinear, the network output is linear for the adjustable parameters, and the weight of the network can be directly solved by a linear equation set, so that the learning speed is greatly increased, and the problem of local minima is avoided.
Currently, most centrifugal blood pumps in ECMO operate at constant rotational speeds (Li Guanhua, zhang, zhao Zongkai, et al ECMO pulsatile flow generating devices: 202121333914.1[ P ]. 2021-12-17). Centrifugal blood pumps operating at constant rotational speeds do not meet the requirements of patients under different active and pathological conditions (e.g. hypertension, hypotension, hypervolemia, hypovolemia, etc.) for different cardiac output (PATIBANLLA P K, RAJASEKARAN N S, SHELAR S B, et al. Evaluation of the effect of diminished pulsatility as seen in continuous flow ventricular assist devices on arterial endothelial cell phenotype and function [ J ]. The Journal of Heart and Lung Transplantation,2016, 35 (7): 930-932.DOI:10.1016/J. Health. 2016.03.008.). More importantly, the constant rotational speed of the working centrifugal blood pump does not produce significant pulsatile flow stimulation to the blood vessel, resulting in significant reduction of pulsatility of the blood vessel (SOUCY K G, KOENIG S C, GIRIDHARAN G A, et a1.Rotary pumps and diminished pulsatility: do we need a pulse [ J ]. ASAIO Journal,2013, 59 (4): 355-366.DOI:10.1097/MAT.0b013e31829f9 bb3.). Various complications caused by reduced vascular pulsatility of constant rotational speed centrifugal blood pumps have been reported clinically, such as arteriovenous malformations, gastrointestinal bleeding, hemorrhagic stroke, aortic insufficiency and valve fusion (SOUCY K G, KOENIG S C, GIRIDHARAN G A, et a1.Defining pulsatility during continuous-flow ventricular assist device support [ J ]. The Journal of Heart and Lung Transplantation,2013, 32 (6): 581-587.DOI:10.1016/J. Health.2013.02.010.) and various neurological complications (Xi Shaosong, zhu Ying, asparagus Meng Yuan, etc. receiving venous arterial adventitial pulmonary oxygenation following cardiac arrest supports the patient' S neurological complications analysis [ J ]. Chinese modern doctor, 2020, 58 (28): 34-40.), and the like.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide an ECMO centrifugal blood pump flow pulsatility control system based on an RBF neural network, which approximates nonlinear terms in an ideal control law through the RBF neural network, so as to realize the control of the ECMO blood pump flow pulsatility and further realize the accurate control of the aortic blood pressure waveform of a patient.
An ECMO (electronic control unit) centrifugal blood pump flow pulsatility control system based on an RBF neural network realizes the pulsatility of the centrifugal blood pump blood flow by periodically changing the rotating speed of the centrifugal blood pump, and comprises a main controller module, a signal acquisition module, a motor driving module and a man-machine interaction module; the main controller module controls the centrifugal blood pump by using an adaptive control algorithm based on the RBF neural network, and calculates the rotating speed of the centrifugal blood pump according to the actual and expected arterial pressures of the patient.
The self-adaptive control algorithm based on the RBF neural network adopts a radial basis function network with a self-adaptive learning function.
The mode of periodically changing the rotating speed of the blood pump is regulated according to a preset blood pressure curve or a preset periodic function describing blood pressure.
The self-adaptive control algorithm based on the RBF neural network predicts the corresponding rotational speed of the centrifugal blood pump according to the actual arterial pressure of the patient so as to maintain the corresponding aortic blood pressure pulsatility.
The ECMO (electronic control unit) centrifugal blood pump flow pulsatility control system based on the RBF neural network has an automatic regulation function, and can automatically regulate the rotating speed of the centrifugal blood pump according to the physiological state and the requirement of a patient so as to provide corresponding blood pulsatility.
The method for establishing the self-adaptive control algorithm based on the RBF neural network comprises the following steps:
1) Establishing a mathematical model of a human blood circulation system;
the dynamic characteristics of the human blood circulation system are simulated by using an electric network, the voltage in the electric network corresponds to blood in the blood circulation system, the current corresponds to blood flow in the blood circulation system, the resistance corresponds to blood flow resistance in the blood circulation system, the capacitance corresponds to vascular compliance in the blood circulation system, the inductance corresponds to blood flow inertia in the blood circulation system, the diode corresponds to a valve in the circulation system, the contraction of the left ventricle and the right ventricle is simulated by using a combined circuit of a controllable voltage source and a controllable current source, and the mathematical model of the obtained blood circulation system is as follows:
the meaning of each variable in the equation is as follows: x is x 1 : aortic blood pressure; x is x 2 : a volumetric flow of blood into the systemic arterial system; x is x 3 : systemic inlet blood pressure; x is x 4 : a volumetric flow of blood into the systemic venous system; x is x 5 : systemic venous system and right atrial inlet blood pressure; x is x 6 : left ventricular volume; x is x 7 : pulmonary arterial blood pressure; x is x 8 : the volumetric flow of blood into the pulmonary circulatory arterial system; x is x 9 : pulmonary blood pressure; x is x 10 : a volumetric flow of blood into the pulmonary circulatory venous system; x is x 11 : pulmonary vein and left atrial partial blood pressure; x is x 12 : left ventricular volume; r is R 1 : aortic valve resistance; r is R 2 : aortic and systemic arterial system resistance; r is R 3 : capillary and systemic venous partial resistance; r is R 4 : % tricuspid resistance; r is R 5 : % pulmonary valve resistance; r is R 6 : pulmonary artery resistance; r is R 7 : other resistance of the lungs; r is R 8 : mitral valve resistance; r is R L : left ventricular internal viscous drag; r is R R : right ventricular internal viscous drag; l (L) 1 : aortic-systemic arterial inertia; l (L) 2 : systemic circulation meridian system inertia; l (L) 3 : pulmonary artery inertia; l (L) 4 : pulmonary vein inertia; c (C) 1 : aortic compliance;C 2 : systemic arterial system compliance; c (C) 3 : systemic venous system compliance; c (C) 4 : pulmonary arterial compliance; c (C) 5 : pulmonary circulatory arterial system compliance; c (C) 6 : pulmonary circulatory venous system compliance; s is S i (t): control valve S i Function of opening and closing, related to the pressure difference across the valve; ΔP i (t): valve S i Differential pressure across;
omitting an equation related to pulmonary circulation in the above equation set; meanwhile, the equation related to ventricular contraction in the equation set is not needed to be considered, so the equation set is simplified into:
2) Establishing a mathematical model of the centrifugal blood pump;
centrifugal blood pumps used in ECMO devices are described using the following equations:
wherein Q represents the blood flow output by the centrifugal blood pump, mountain represents the rotational speed of the centrifugal blood pump, and P out Represents the outlet pressure of the centrifugal blood pump, P in Represents the inlet pressure of the centrifugal blood pump beta 0 、β 1 、β 2 Is a constant, and is obtained by least square fitting according to experimental data of the centrifugal blood pump;
when the constant in the mathematical model of the centrifugal blood pump is determined, the variable m is used 1 、m 2 、m 3 Replacing coefficients in a mathematical model of the centrifugal blood pump, wherein the mathematical model of the centrifugal blood pump is changed into:
centrifugal blood pump-blood circulation coupled system simulation module: combining the mathematical model of the centrifugal blood pump with the mathematical model of the human blood circulation system to obtain a mathematical model of the centrifugal blood pump-blood circulation coupling system;
the inlet of the centrifugal blood pump is directly connected with the junction of the upper and lower vena cava of the patient, and the inlet pressure of the centrifugal blood pump is approximately considered to be the systemic venous system and the right atrium inlet blood pressure in the human blood circulation system, namely P in ＝x 5 The method comprises the steps of carrying out a first treatment on the surface of the A resistor R is added between the centrifugal blood pump and the circulating system in the coupling model p To represent resistance to blood circulation in the oxygenator;
for the centrifugal blood pump-circulatory system coupling model, the new state variables are selected as follows:
z 1 : aortic blood pressure; z 2 : the current in L1, the volumetric flow of blood into the systemic arterial system; z 3 : voltage on C2, systemic inlet blood pressure; z 4 : the current in L2, the volumetric flow of blood into the systemic venous system; z 5 : voltage on C3, systemic venous system and right atrial blood pressure; z 6 : centrifugal blood pump flow;
let input u be the square of the rotational speed of the centrifugal blood pump and output y be the aortic blood pressure, the state equation of the centrifugal blood pump-circulatory system coupling model can be obtained as follows:
the output equation of the centrifugal blood pump-circulatory system coupling model is as follows:
y＝z 1 。
3) Designing a control law of a system and a weight update law of an RBF neural network based on Lyapunov stability theory;
the state equation of the centrifugal blood pump-circulatory system coupling model can be known:
again toThe derivation can be obtained:
order theWherein->The above can be written as:
the tracking error of the system is as follows:
e＝y-y d ＝z 1 -z 1d
wherein y is d And z 1d Is the desired aortic blood pressure;
defining a tracking error function:
wherein lambda > 0, it is easy to see that when s.fwdarw.0, there is e.fwdarw.0 andtherefore, a control law is designed to ensure that the error function s is gradually stabilized at the s=0 point, so that the tracking error of the centrifugal blood pump-circulatory system coupling model control is ensured to be gradually stabilized at the e=0 point;
constructing a Lyapunov function:
the method can obtain:
order theObtaining:
according to Lyapunov stability theory, it is required to makeNegative determination, can make:
the ideal controller can be obtained by:
wherein eta is a constant greater than zero for controllingA distance from zero; selecting RBF neural network approximation function +.>
The RBF neural network is divided into an input layer, an hidden layer and an output layer, wherein the expression of an activation function of the hidden layer neuron is as follows:
where x is the input vector of the neuron, c j Center position for the jth neuron activation function, b j The width of the activation function for the jth neuron;
when using RBF neural network approximation function f (x), there must be an ideal weight vector W * Such that:
f(x)＝W *T h(x)+∈
wherein W is * Is an ideal weight vector of the neural network, and E is an approximation error of the neural network, and E is < [ E ] N ，∈ N Is any real number greater than zero; approximation function using RBF neural networkLet the actual output of the neural network be:
as an actual weight vector of the RBF neural network, an approximation error of the neural network can be expressed as:
wherein the weight approximates the errorDesign the adaptive law such that +.>The value of (2) is +.>Progressive stabilization of the position; the Lyapunov function describing both tracking error and approximation error is designed:
wherein gamma is more than 0, and the following steps are obtained:
firstly, the tracking error is ensured to be stabilized to 0 gradually, and the design control law is as follows:
when η > |e| max Time s epsilon-s eta sgn(s) is negatively set, and the adaptive law is taken:
obtaining:
when η > |e| max Time of dayNegative determination, the whole control system is gradually stable at the position where the tracking error and the network approximation error approach zero;
compared with the prior art, the invention has the beneficial effects that:
according to the invention, the pulsatility of the blood flow of the blood pump is realized by periodically changing the rotating speed of the blood pump, so that the pulsatility of the aortic blood pressure of a patient using ECMO is improved; meanwhile, the invention adopts a control algorithm comprising a radial basis function neural network with a self-adaptive learning function, so that the invention can automatically adjust the rotating speed of the blood pump according to the physiological state and the requirement of a patient so as to provide proper blood flow pulsatility.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention.
Fig. 2 is a schematic diagram of an RBF neural network according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a model of a human blood circulation network constructed in an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the drawings and examples.
As shown in fig. 1, an ECMO centrifugal blood pump flow pulsatility control system based on RBF neural network, which realizes pulsatility of blood flow of the centrifugal blood pump by periodically changing rotational speed of the centrifugal blood pump, thereby improving aortic blood pressure pulsatility of a patient, includes: the system comprises a main controller module, a signal acquisition module, a motor driving module and a man-machine interaction module;
the main controller module controls the centrifugal blood pump by using a self-adaptive control algorithm based on the RBF neural network, and calculates the rotating speed of the centrifugal blood pump according to the actual and expected arterial pressures of the patient; the self-adaptive control algorithm based on the RBF neural network adopts a radial basis function network with a self-adaptive learning function, and predicts the proper rotational speed of the centrifugal blood pump according to the actual aortic pressure of a patient so as to maintain proper aortic blood pressure pulsatility.
The system has an automatic regulation function, and can automatically regulate the rotating speed of the centrifugal blood pump according to the physiological state and the requirement of a patient so as to provide proper blood flow pulsatility.
The main controller module adopts ZYNQ series chips, and the specific model is XC7Z020CLG400-2;
the signal acquisition module is used for acquiring the arterial pressure, the centrifugal blood pump flow and other related parameters of a patient in real time by installing pressure sensors in the ECMO arterial cannula and at the inlet of the ECMO centrifugal blood pump and installing flow sensors on the outlet pipeline of the centrifugal blood pump;
the motor driving module uses a three-phase half-bridge driving circuit to drive a brushless direct current motor connected with the centrifugal blood pump;
the man-machine interaction module comprises a touch screen and an alarm system, wherein the touch screen is used for displaying important operation parameters, operation states and alarm information of the system and inputting control instructions; the alarm system is used for monitoring abnormal rotation speed of the centrifugal blood pump and abnormal aortic blood pressure of a patient and sending out corresponding alarms.
The core of the system is an adaptive control algorithm based on an RBF neural network, the actual arterial pressure of a patient and the expected arterial pressure are taken as inputs, and the current rotational speed of the centrifugal blood pump is calculated and obtained through the RBF neural network; the RBF neural network structure diagram in the self-adaptive control algorithm based on the RBF neural network is shown in figure 2, and the RBF neural network is characterized in that an input layer and an hidden layer are connected without weights, an input signal is directly mapped to a hidden layer space through a radial basis function, so that mapping of the signal from low dimension to high dimension is realized, the input signal is linearly separable in the high dimension, the mapping is nonlinear mapping, the mapping relation is only related to radial basis function parameters, and the adjustment of the input mapping can be realized by adjusting the radial basis function parameters; but the hidden layer and the output layer are connected by weight, namely the mapping from the hidden layer space to the output space is linear; the mapping from input to output of the whole neural network is nonlinear, the network output is linear for the adjustable parameters, and the weight of the network can be directly solved by a linear equation set, so that the learning speed is greatly increased, and the problem of local minima is avoided.
The method for establishing the self-adaptive control algorithm based on the RBF neural network comprises the following steps:
1) The human body blood circulation network is simulated by using an electric network, the voltage in the electric network corresponds to blood in a blood circulation system, the current corresponds to blood flow in the blood circulation system, the resistance corresponds to blood flow resistance in the blood circulation system, the capacitance corresponds to vascular compliance in the blood circulation system, the inductance corresponds to blood flow inertia in the blood circulation system, the diode corresponds to a valve in the circulation system, the contraction of a left ventricle and a right ventricle is simulated by using a combined circuit of a controllable voltage source and a controllable current source, and the mathematical model of the obtained blood circulation system is as follows as shown in fig. 3:
the meaning of each variable in the equation is as follows: x is x 1 : aortic blood pressure; x is x 2 : a volumetric flow of blood into the systemic arterial system; x is x 3 : systemic inlet blood pressure; x is x 4 : a volumetric flow of blood into the systemic venous system; x is x 5 : systemic venous system and right atrial inlet blood pressure; x is x 6 : left ventricular volume; x is x 7 : pulmonary arterial blood pressure; x is x 8 : the volumetric flow of blood into the pulmonary circulatory arterial system; x is x 9 : pulmonary blood pressure; x is x 10 : a volumetric flow of blood into the pulmonary circulatory venous system; x is x 11 : pulmonary vein and left atrial partial blood pressure; x is x 12 : left ventricular volume; r is R 1 : aortic valve resistance; r is R 2 : aortic and systemic arterial system resistance; r is R 3 : capillary and systemic venous partial resistance; r is R 4 : % tricuspid resistance; r is R 5 : % pulmonary valve resistance; r is R 6 : pulmonary artery resistance; r is R 7 : other resistance of the lungs; r is R 8 : mitral valve resistance; r is R L : left ventricular internal viscous drag; r is R R : right ventricular internal viscous drag; l (L) 1 : aortic-systemic arterial inertia; l (L) 2 : systemic circulation meridian system inertia; l (L) 3 : pulmonary artery inertia; l (L) 4 : pulmonary vein inertia; c (C) 1 : aortic compliance; c (C) 2 : systemic arterial system compliance; c (C) 3 : systemic venous system compliance; c (C) 4 : pulmonary arterial compliance; c (C) 5 : pulmonary circulatory arterial system compliance; c (C) 6 : pulmonary circulatory venous system compliance; s is S i (t): control valve S i Function of opening and closing, related to the pressure difference across the valve; ΔP i (t): valve S i Differential pressure across;
considering the application scenario of ECMO, when the patient adapts to ECMO, no blood flows through the pulmonary circulation system, so that the equation related to pulmonary circulation in the above equation set can be omitted; meanwhile, under the condition of applying VA-ECMO, ECMO directly pumps blood from the junction of the upper and lower vena cava to the aorta, and the contraction and the relaxation of the heart hardly directly affect the dynamics of the circulatory system, so that the equation related to ventricular contraction in the equation set is not needed to be considered, and the equation set can be simplified into:
2) Establishing a mathematical model of the centrifugal blood pump;
centrifugal blood pumps used in ECMO devices are described using the following equations:
wherein Q represents the blood flow output by the centrifugal blood pump, mountain represents the rotational speed of the centrifugal blood pump, and P out Represents the outlet pressure of the centrifugal blood pump, P in Represents the inlet pressure of the centrifugal blood pump beta 0 、β 1 、β 2 Is a constant and can be obtained by least square fitting according to experimental data of the centrifugal blood pump;
when the constant in the mathematical model of the centrifugal blood pump is determined, the variable m is used 1 、m 2 、m 3 Replacing coefficients in a mathematical model of the centrifugal blood pump, wherein the mathematical model of the centrifugal blood pump is changed into:
centrifugal blood pump-blood circulation coupled system simulation module: combining the mathematical model of the centrifugal blood pump with the mathematical model of the human blood circulation system to obtain a mathematical model of the centrifugal blood pump-blood circulation coupling system;
considering that the inlet of the centrifugal blood pump can be considered to be directly connected with the junction of the upper and lower vena cava of the patient, the inlet pressure of the centrifugal blood pump can be approximately considered to be the systemic venous system and the right atrium inlet blood pressure in the human blood circulation system, namely P in ＝x 5 The method comprises the steps of carrying out a first treatment on the surface of the For the outlet of the centrifugal blood pump, consider the outlet of the centrifugal blood pumpAn oxygenator is arranged between the blood pump and the aorta of the patient, so that a resistor R is added between the centrifugal blood pump and the circulatory system in the coupling model p To represent resistance to blood circulation in the oxygenator; for the centrifugal blood pump-circulatory system coupling model, the new state variables are selected as follows:
z 1 : aortic blood pressure; z 2 : the current in L1, the volumetric flow of blood into the systemic arterial system; z 3 : voltage on C2, systemic inlet blood pressure; z 4 : the current in L2, the volumetric flow of blood into the systemic venous system; z 5 : voltage on C3, systemic venous system and right atrial blood pressure; z 6 : centrifugal blood pump flow;
let input u be the square of the rotational speed of the centrifugal blood pump and output y be the aortic blood pressure, the state equation of the centrifugal blood pump-circulatory system coupling model can be obtained as follows:
the output equation of the centrifugal blood pump-circulatory system coupling model is as follows:
y＝z 1
3) Designing a control law of a system and a weight update law of an RBF neural network based on Lyapunov stability theory;
the state equation of the centrifugal blood pump-circulatory system coupling model can be known:
again toThe derivation can be obtained:
order theWherein->The above can be written as:
the tracking error of the system is as follows:
e＝y-y d ＝z 1 -z 1d
wherein y is d And z 1d Is the desired aortic blood pressure;
defining a tracking error function:
wherein lambda > 0, it is easy to see that when s.fwdarw.0, there is e.fwdarw.0 andtherefore, a control law is designed to ensure that the error function s is gradually stabilized at the s=0 point, so that the tracking error of the centrifugal blood pump-circulatory system coupling model control is ensured to be gradually stabilized at the e=0 point;
constructing a Lyapunov function:
the method can obtain:
order theObtaining:
according to Lyapunov stability theory, it is required to makeNegative determination, can make:
the ideal controller can be obtained by:
wherein eta is a constant greater than zero for controllingA distance from zero; in the above formula, the function->The degree of nonlinearity of (2) is high and the expression is difficult to be obtained, so the RBF neural network approximation function is selected to be used +.>
In the RBF neural network used in the invention, the input layer is provided with 5 neurons, the hidden layer is provided with 21 neurons, and the output layer is provided with 1 neuron, wherein the general expression of the activation function of the hidden layer neurons is as follows:
where x is the input vector of the neuron, c j Center position for the jth neuron activation function, b j The width of the activation function for the jth neuron;
theoretically, the RBF neural network can approach a nonlinear function with arbitrary accuracy, so that when the RBF neural network is used to approach the function f (x), there must be an ideal weight vector W * Such that:
f(x)＝W *T h(x)+∈
wherein W is * Is an ideal weight vector of the neural network, and E is an approximation error of the neural network, and E is < [ E ] N ，∈ N Is any real number greater than zero; thus, an RBF neural network approximation function can be usedLet the actual output of the neural network be:
as an actual weight vector of the RBF neural network, an approximation error of the neural network can be expressed as:
wherein the weight approximates the errorDesign the adaptive law such that +.>The value of (2) is +.>Progressive stabilization of the position; to stabilize the control globally, a Lyapunov function is designed that describes both tracking error and approximation error: />
Wherein γ > 0, obtainable:
firstly, the tracking error is ensured to be stabilized to 0 gradually, and the design control law is as follows:
when η > |e| max Time s epsilon-s eta sgn(s) is negatively set, and the adaptive law is taken:
the method can obtain:
it can be known that when η > |e| max Time of dayNegative determination, the whole control system is gradually stable at the position where the tracking error and the network approximation error approach zero;
the self-adaptive control algorithm based on the RBF neural network operates in the main controller module as follows:
1) Initializing; setting initial operation parameters based on RBF neural network, wherein the operation parameters comprise: center position matrix c of radial basis function in RBF neural network ij Width vector b of radial basis function j The weight vector W of the RBF neural network, a constant lambda in an error function and an adaptive gain matrix T;
2) Running an adaptive control algorithm based on an RBF neural network; the preset RBF neural network calculates the control quantity of the current centrifugal blood pump-blood circulation coupling system according to the current weight and the network input parameters, and meanwhile, the weight update law of the preset RBF neural network corrects the weight vector of the RBF neural network in real time according to the current tracking error, so that the tracking error is reduced as much as possible.
Claims (10)
1. An ECMO (electronic control unit) centrifugal blood pump flow pulsatility control system based on an RBF neural network is characterized by realizing the pulsatility of the centrifugal blood pump blood flow by periodically changing the rotating speed of the centrifugal blood pump, and comprises a main controller module, a signal acquisition module, a motor driving module and a man-machine interaction module; the main controller module controls the centrifugal blood pump by using an adaptive control algorithm based on the RBF neural network, and calculates the rotating speed of the centrifugal blood pump according to the actual and expected arterial pressures of the patient.
2. The system according to claim 1, wherein: the self-adaptive control algorithm based on the RBF neural network adopts a radial basis function network with a self-adaptive learning function.
3. The system according to claim 1, wherein: the mode of periodically changing the rotating speed of the blood pump is regulated according to a preset blood pressure curve or a preset periodic function describing blood pressure.
4. The system according to claim 1, wherein: the self-adaptive control algorithm based on the RBF neural network predicts the corresponding rotational speed of the centrifugal blood pump according to the actual arterial pressure of the patient so as to maintain the corresponding aortic blood pressure pulsatility.
5. The system according to claim 1, wherein: the automatic blood flow regulating device has an automatic regulating function, and can automatically regulate the rotating speed of the centrifugal blood pump according to the physiological state and the requirement of a patient so as to provide corresponding blood flow pulsatility.
6. The system according to claim 1, wherein: the main controller module adopts ZYNQ series chips, and the model is XC7Z020CLG400-2.
7. The system according to claim 1, wherein: the data acquisition module is used for acquiring the arterial pressure, the centrifugal blood pump flow and other relevant parameters of a patient in real time by installing pressure sensors in the ECMO arterial cannula and at the inlet of the ECMO centrifugal blood pump and installing flow sensors on the outlet pipeline of the centrifugal blood pump.
8. The system according to claim 1, wherein: the motor driving module uses a three-phase half-bridge driving circuit to drive a brushless direct current motor connected with the centrifugal blood pump.
9. The system according to claim 1, wherein: the man-machine interaction module comprises a touch screen and an alarm system, wherein the touch screen is used for displaying operation parameters, operation states and alarm information of the system and inputting control instructions; the alarm system is used for monitoring abnormal rotation speed of the blood pump and abnormal aortic blood pressure of the patient and giving out corresponding alarms.
10. The system of claim 3, wherein the method for establishing the RBF-based adaptive control algorithm comprises the following steps:
1) The human body blood circulation network is simulated by using an electric network, voltage in the electric network corresponds to blood in a blood circulation system, current corresponds to blood flow in the blood circulation system, resistance corresponds to blood flow resistance in the blood circulation system, capacitance corresponds to vascular compliance in the blood circulation system, inductance corresponds to blood flow inertia in the blood circulation system, diode corresponds to a valve in the circulation system, and contraction of left and right ventricles is simulated by using a combined circuit of a controllable voltage source and a controllable current source, so that the mathematical model of the obtained blood circulation system is as follows:
the meaning of each variable in the equation is as follows: x is x 1 : aortic blood pressure; x is x 2 : a volumetric flow of blood into the systemic arterial system; x is x 3 : systemic inlet blood pressure; x is x 4 : a volumetric flow of blood into the systemic venous system; x is x 5 : systemic venous system and right atrial inlet blood pressure; x is x 6 : left ventricular volume; x is x 7 : pulmonary arterial blood pressure; x is x 8 : the volumetric flow of blood into the pulmonary circulatory arterial system; x is x 9 : pulmonary blood pressure; x is x 10 : a volumetric flow of blood into the pulmonary circulatory venous system; x is x 11 : pulmonary vein and left atrial partial blood pressure; x is x 12 : left ventricular volume; r is R 1 : aortic valve resistance; r is R 2 : aortic and systemic arterial system resistance; r is R 3 : capillary and systemic venous partial resistance; r is R 4 : % tricuspid resistance; r is R 5 : % pulmonary valve resistance; r is R 6 : pulmonary artery resistance; r is R 7 : other resistance of the lungs; r is R 8 : mitral valve resistance; r is R L : left ventricular internal viscous drag; r is R R : right ventricular internal viscous drag; l (L) 1 : aortic-systemic arterial inertia; l (L) 2 : systemic circulation meridian system inertia; l (L) 3 : pulmonary artery inertia; l (L) 4 : pulmonary vein inertia; c (C) 1 : aortic compliance; c (C) 2 : systemic arterial system compliance; c (C) 3 : systemic venous system compliance; c (C) 4 : pulmonary arterial compliance; c (C) 5 : pulmonary circulatory arterial system compliance; c (C) 6 : pulmonary circulatory venous system compliance; s is S i (t): control valve S i Function of opening and closing, related to the pressure difference across the valve; ΔP i (t): valve S i Differential pressure across;
omitting an equation related to pulmonary circulation in the above equation set; meanwhile, the equation related to ventricular contraction in the equation set is not needed to be considered, and the equation set is simplified into:
2) Establishing a mathematical model of the centrifugal blood pump;
centrifugal blood pumps used in ECMO devices are described using the following equations:
wherein Q represents the blood flow output by the centrifugal blood pump, mountain represents the rotational speed of the centrifugal blood pump, and P out Represents the outlet pressure of the centrifugal blood pump, P in Represents the inlet pressure of the centrifugal blood pump beta 0 、β 1 、β 2 Is a constant, and is obtained by least square fitting according to experimental data of the centrifugal blood pump;
when the constant in the mathematical model of the centrifugal blood pump is determined, the variable m is used 1 、m 2 、m 3 Replacing coefficients in a mathematical model of the centrifugal blood pump, wherein the mathematical model of the centrifugal blood pump is changed into:
the inlet of the centrifugal blood pump is directly connected with the junction of the upper and lower vena cava of the patient, and the inlet pressure of the centrifugal blood pump is considered to be the systemic venous system and the right atrium inlet blood pressure in the human blood circulation system, namely P in ＝x 5 The method comprises the steps of carrying out a first treatment on the surface of the A resistor R is added between the centrifugal blood pump and the circulating system in the coupling model p To represent resistance to blood circulation in the oxygenator;
for the centrifugal blood pump-circulatory system coupling model, the new state variables are selected as follows:
z 1 : aortic blood pressure; z 2 : the current in L1, the volumetric flow of blood into the systemic arterial system; z 3 : voltage on C2, systemic inlet blood pressure; z 4 : current in L2, flowA volumetric flow of blood into the systemic venous system; z 5 : voltage on C3, systemic venous system and right atrial blood pressure; z 6 : centrifugal blood pump flow;
let input u be the square of the rotational speed of the centrifugal blood pump and output y be the aortic blood pressure, the state equation of the centrifugal blood pump-circulatory system coupling model is obtained as follows:
the output equation of the centrifugal blood pump-circulatory system coupling model is as follows:
y＝z 1
3) Designing a control law of a system and a weight update law of an RBF neural network based on Lyapunov stability theory;
from the state equation of the centrifugal blood pump-circulatory system coupling model, it is known that:
again toAnd (3) deriving:
order theWherein->The above-mentioned writing:
the tracking error of the system is as follows:
e＝y-y d ＝z 1 -z 1d
wherein y is d And z 1d Is the desired aortic blood pressure;
defining a tracking error function:
wherein lambda > 0, when s.fwdarw.0, there is e.fwdarw.0 andso a control law is designed to ensure that an error function s is gradually stabilized at s=0, namely the tracking error controlled by the centrifugal blood pump-circulatory system coupling model is ensured to be gradually stabilized at e=0;
constructing a Lyapunov function:
obtaining:
order theObtaining:
according to Lyapunov stability theory, it is required to makeNegative setting, and making:
the ideal controller is obtained by the following steps:
wherein eta is a constant greater than zero for controllingA distance from zero; selecting RBF neural network approximation function
The RBF neural network is divided into an input layer, an hidden layer and an output layer, wherein the expression of an activation function of the hidden layer neuron is as follows:
where x is the input vector of the neuron, c j Center position for the jth neuron activation function, b j The width of the activation function for the jth neuron;
when using RBF neural network approximation function f (x), there must be an ideal weight vector W * Such that:
f(x)＝W *T h(x)+∈
wherein W is * Is an ideal weight vector of the neural network, and E is an approximation error of the neural network, and E is < [ E ] N ，∈ N Is any real number greater than zero; approximation function using RBF neural networkLet the actual output of the neural network be:
as an actual weight vector of the RBF neural network, an approximation error of the neural network is expressed as:
wherein the weight approximates the errorDesign the adaptive law such that +.>The value of (2) is +.>Progressive stabilization of the position; the Lyapunov function describing both tracking error and approximation error is designed:
wherein gamma is more than 0, and the following steps are obtained:
firstly, the tracking error is ensured to be stabilized to 0 gradually, and the design control law is as follows:
when η > |e| max Time s epsilon-s eta sgn(s) is negatively set, and the adaptive law is taken:
obtaining:
when η > |e| max Time of dayNegative determination is carried out, and the whole control system is gradually stable at the position where the tracking error and the network approximation error are close to zero.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310742773.6A CN116650827A (en) | 2023-06-21 | 2023-06-21 | Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310742773.6A CN116650827A (en) | 2023-06-21 | 2023-06-21 | Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116650827A true CN116650827A (en) | 2023-08-29 |
Family
ID=87713716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310742773.6A Pending CN116650827A (en) | 2023-06-21 | 2023-06-21 | Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116650827A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117727463A (en) * | 2023-12-28 | 2024-03-19 | 中国药科大学 | A QSP model based on the physiological mechanism of blood circulation and its application |
| CN119846956A (en) * | 2024-12-23 | 2025-04-18 | 上海交通大学 | Ultrasonic actuation pump lung control method based on reinforcement learning and fuzzy control |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107045281A (en) * | 2016-12-20 | 2017-08-15 | 江苏大学 | A kind of variable speed control method based on heart pump new model |
| CN109641089A (en) * | 2016-08-23 | 2019-04-16 | 阿比奥梅德欧洲股份有限公司 | Ventricular assist device |
| CN111481763A (en) * | 2020-04-24 | 2020-08-04 | 南京鼓楼医院 | Pulse artificial lung cardiopulmonary bypass manager |
| CN112891728A (en) * | 2021-02-26 | 2021-06-04 | 佛山科学技术学院 | Self-adaptive control blood pump supply system and method |
| CN113476737A (en) * | 2021-07-02 | 2021-10-08 | 中国科学院江西稀土研究院 | Method, device and storage medium for improving bionic pulsation of artificial heart pump blood flow |
| CN215194330U (en) * | 2021-06-16 | 2021-12-17 | 中山大学孙逸仙纪念医院 | ECMO pulsating flow generation device |
| CN114423486A (en) * | 2019-07-10 | 2022-04-29 | 伊克尼奥斯股份有限公司 | Control for non-occlusive blood pumps |
| CN115869529A (en) * | 2022-12-15 | 2023-03-31 | 深圳市博迈本尚医疗科技有限公司 | Blood pumping device and ECMO equipment |
| CN116212138A (en) * | 2022-12-27 | 2023-06-06 | 卫圣康医学科技(江苏)有限公司 | Method for establishing bidirectional feedback regulation between ECMO and blood pressure monitoring equipment |
| CN116235846A (en) * | 2023-02-27 | 2023-06-09 | 中国科学院苏州生物医学工程技术研究所 | Organ perfusion control method, system and device |
-
2023
- 2023-06-21 CN CN202310742773.6A patent/CN116650827A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109641089A (en) * | 2016-08-23 | 2019-04-16 | 阿比奥梅德欧洲股份有限公司 | Ventricular assist device |
| CN107045281A (en) * | 2016-12-20 | 2017-08-15 | 江苏大学 | A kind of variable speed control method based on heart pump new model |
| CN114423486A (en) * | 2019-07-10 | 2022-04-29 | 伊克尼奥斯股份有限公司 | Control for non-occlusive blood pumps |
| CN111481763A (en) * | 2020-04-24 | 2020-08-04 | 南京鼓楼医院 | Pulse artificial lung cardiopulmonary bypass manager |
| CN112891728A (en) * | 2021-02-26 | 2021-06-04 | 佛山科学技术学院 | Self-adaptive control blood pump supply system and method |
| CN215194330U (en) * | 2021-06-16 | 2021-12-17 | 中山大学孙逸仙纪念医院 | ECMO pulsating flow generation device |
| CN113476737A (en) * | 2021-07-02 | 2021-10-08 | 中国科学院江西稀土研究院 | Method, device and storage medium for improving bionic pulsation of artificial heart pump blood flow |
| CN115869529A (en) * | 2022-12-15 | 2023-03-31 | 深圳市博迈本尚医疗科技有限公司 | Blood pumping device and ECMO equipment |
| CN116212138A (en) * | 2022-12-27 | 2023-06-06 | 卫圣康医学科技(江苏)有限公司 | Method for establishing bidirectional feedback regulation between ECMO and blood pressure monitoring equipment |
| CN116235846A (en) * | 2023-02-27 | 2023-06-09 | 中国科学院苏州生物医学工程技术研究所 | Organ perfusion control method, system and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117727463A (en) * | 2023-12-28 | 2024-03-19 | 中国药科大学 | A QSP model based on the physiological mechanism of blood circulation and its application |
| CN119846956A (en) * | 2024-12-23 | 2025-04-18 | 上海交通大学 | Ultrasonic actuation pump lung control method based on reinforcement learning and fuzzy control |
| CN119846956B (en) * | 2024-12-23 | 2025-12-12 | 上海交通大学 | Ultrasonic actuation pump lung control method based on reinforcement learning and fuzzy control |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5098370A (en) | Heart assist device | |
| Boston et al. | Hierarchical control of heart-assist devices | |
| EP1942965B1 (en) | Total artificial heart system for auto-regulating flow and pressure balance | |
| US20050159639A1 (en) | Physiologically based control system and method for using the same | |
| CN116650827A (en) | Flow Pulsatility Control System of ECMO Centrifugal Blood Pump Based on RBF Neural Network | |
| Frazier et al. | Optimization of axial-pump pressure sensitivity for a continuous-flow total artificial heart | |
| CN113476737B (en) | Method, device and storage medium for improving bionic pulsatility of blood flow of artificial heart pump | |
| CN109876216B (en) | LVAD (Low Voltage differential analog) undifferentiated self-adaptive fuzzy PI (proportional integral) control method based on noninvasive measurement | |
| CN115645735B (en) | Multi-target physiological control method, system and device for artificial heart pump and storage medium | |
| Bozkurt et al. | Enhancement of arterial pressure pulsatility by controlling continuous-flow left ventricular assist device flow rate in mock circulatory system | |
| CN116235846B (en) | Organ perfusion control method, system and device | |
| Cordeiro et al. | A physiological control system for ECG-synchronized pulsatile pediatric ventricular assist devices | |
| Stevens et al. | Physiological control | |
| CN118105617A (en) | Multi-target self-adaptive control method for heart pump based on Frank-starling mechanism | |
| CN116421877B (en) | A control device, ventricular assist system and adaptive control method | |
| Ketelhut et al. | Iterative learning control of a left ventricular assist device | |
| Koh et al. | A centralized multi-objective model predictive control for a biventricular assist device: An in vitro evaluation | |
| CN119345597A (en) | Ventricular assist device control device, method, ventricular assist system and electronic device | |
| Melo et al. | Feedback controller for restoring the basal hemodynamic condition with a rotary blood pump used as left ventricular assist device | |
| CN120037574A (en) | Ventricular assist device rotating speed regulation and control method and system based on deep learning | |
| Antaki et al. | Control of heart assist devices | |
| Yoshizawa et al. | An automatic control algorithm for the optimal driving of the ventricular-assist device | |
| CN118924476A (en) | In vitro organ perfusion control method, system and device for experimental small animals | |
| CN116510171A (en) | Method and system for controlling driving of nerve pulsation of artificial heart | |
| Rubtsova et al. | Lumped parameter model of the systemic circulation and a rotary blood pump interaction |
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 |