CN103007391B - Predictive adaptive control system for liquid medicament push-injecting - Google Patents

Abstract

The invention discloses a predictive self-adaptive control system for drug liquid injection, and relates to the technical field of automatic control of drug injection in clinical operations. The system includes a main control module and a drug injection module, the drug injection module includes a microcontroller and a medical syringe, the microcontroller is connected to the predictive control algorithm module of the main control module, and the predictive control algorithm module controls the time of the next cardiac cycle The bolus injection time is sent to the microcontroller, and the microcontroller controls the micro propeller to control the advancing speed, bolus injection time and bolus injection stop time of the medical syringe. In the present invention, the medicine pushing speed, the time of injection, the time of stopping the injection and the amount of pushing medicine in the Chinese medicine liquid injection module are controlled by the main control module, which improves the reaction speed and real-time control ability of the medicine liquid injection module, and realizes the Scientific, rational, and safe medication, that is, with a lower drug dose, obtain the maximum drug concentration in the local organs and stay for a long enough time without reflux, reducing damage to healthy tissues.

Description

Predictive Adaptive Control System for Bolus Injection

technical field

The invention relates to the technical field of automatic control of medicine pushing in clinical operations, in particular to a predictive self-adaptive control system for medicine liquid injection. the

Background technique

At present, in the clinical treatment of cancer, chemotherapy through vascular intervention is commonly used, that is, puncture and catheterization under the monitoring of digital subtraction X-ray machine (DSA), and the catheter is introduced into the entrance of the branch of the tumor blood supply artery. , inject chemotherapy drugs at the inlet end of the catheter. This method was first used for cardiovascular diseases, and was soon applied to the diagnosis and treatment of various diseases in various parts of the body, such as the nervous system and tumors. In my country, interventional therapy is especially used for tumors, accounting for the majority of interventional therapy, and has become a commonly used therapy in comprehensive tumor therapy. The intervention methods also range from vascular intervention to pulmonary-pulmonary intervention, bronchial artery intervention, intracavitary intervention, and direct intratumoral intervention. the

At present, in the clinical practice of pushing drugs through blood vessels, it is common to only focus on tumor lesions and cancer cells. Therefore, aggressive drug pushing chemotherapy with large doses and high-speed perfusion is adopted with the goal of destroying cancer cells with a simple hand-push syringe. However, when using this traditional method of manually pushing the medicine, the medicine liquid will follow the blood flow direction in the blood vessel to all parts of the body after entering the blood vessel, which cannot be compared with the patient's blood pressure value, heart rate, etc., which will respond to the patient's own adjustment mechanism. The parameters are connected, so that only part of the liquid medicine enters the tumor blood vessels, resulting in the result that the liquid medicine kills both cancer cells and normal cells; It will also cause many clinical problems, which will not only increase the financial burden of patients, but also increase the workload of medical staff. the

Contents of the invention

The purpose of the present invention is to overcome the above-mentioned disadvantages in the prior art, and provide a predictive adaptive control system for drug fluid bolus injection that can reduce damage to healthy tissues. the

In order to realize the above-mentioned purpose of the invention, the present invention provides the following technical solutions:

A predictive adaptive control system for medicinal liquid bolus injection, comprising a main control module and a medicinal liquid bolus injection module, characterized in that the main control module includes a detection module for collecting human blood pressure values, and the detection module Including a blood pressure sensor; the detection module transmits the blood pressure value to the memory area for storage to form an array of blood pressure values, the memory area is connected to an identification module, and the identification module identifies the blood pressure value array in the memory area to obtain real-time cardiac cycle and The characteristic parameters in the cycle, the identification module is also connected with a predictive control algorithm module, and the predictive control algorithm module obtains the injection time of the next cardiac cycle by calculating the characteristic parameters;

The medicinal liquid bolus injection module is connected to the predictive control algorithm module of the main control module, the medicinal liquid bolus injection module includes a microcontroller and a medical syringe, and the medicinal liquid bolus injection module is connected to the predictive control algorithm module through the microcontroller , the predictive control algorithm module sends the bolus injection time of the next cardiac cycle to the microcontroller, the microcontroller is connected with a micro-propeller, the micro-propeller is connected with a medical injector, and the micro-controller controls the micro-propeller The propulsion speed, bolus injection time and bolus injection stop time of the propeller, and then control the injection speed, bolus injection time and bolus injection stop time of the medical syringe;

The system processes signals according to the following steps:

(1) Collection of blood pressure values: real-time collection of human blood pressure values through the detection module in the main control module, and recursively stored in the memory area of the main control module to form an array of blood pressure values;

(2) Obtain relevant parameters: the identification module in the main control module obtains the real-time cardiac cycle and the characteristic parameters in the cycle by identifying the array of blood pressure values; obtains the patient's symptom parameters through physical examination;

(3) Calculating the time to push the medicine: set the time to push the medicine according to the symptom parameters; the main control module calculates the time to push the medicine in the next cardiac cycle through the predictive control algorithm according to the time to push the medicine, the real-time cardiac cycle and the characteristic parameters in the cycle ;

(4) Start the medicinal solution injection module: the main control module sends the calculated drug pushing time in the next cardiac cycle to the microcontroller in the medicinal solution injection module, and the microcontroller calculates the drug pushing time according to the drug pushing time and Ahead of t _delay , start the micro injector in the liquid medicine bolus injection module to carry out the liquid medicine bolus injection, and said t _delay is the time delay of the liquid medicine bolus injection module;

(5) Self-adaptation of bolus injection: the detection module collects new blood pressure values, and compares the newly obtained sampled blood pressure values, newly identified cardiac cycle and characteristic parameters in the cycle with the previously collected blood pressure value, cardiac cycle and the characteristic parameters in the cycle. The weighted average of the characteristic parameters updates the parameters of the predictive control algorithm to realize the adaptability of the predictive control bolus injection of medicinal liquid;

In step (1), the recursive method means that when sampling a new blood pressure value into the blood pressure value array, a group of earliest sampled blood pressure values are discarded at the same time, so that the dimension of the blood pressure value array remains unchanged;

In step (2), the characteristic parameters in the cardiac cycle include systolic blood pressure, diastolic blood pressure, diastolic duration, and systolic duration; symptom parameters include tumor diameter, tumor blood supply artery, distance between the tumor and the preset drug injection point ;

In step (3), the medicine pushing moment refers to the starting moment of a phase in the characteristic parameters of the cardiac cycle; the medicine pushing moment t _r (k+1) in the next cardiac cycle is passed through the equation t _r (k+ 1)'=t _r (k)+T _m (k)+

ξ[(t _r (k)-t _r (k)',t _r (k-1)-t _r (k-1)',Lt _r (km)-t _r (k=m)'] ^T to Calculation, where, t _r (k) is the initial moment of this phase in the current cardiac cycle, that is, the current drug pushing moment; t _r (k)' is calculated by the predictive control algorithm based on the characteristic parameters in the last cardiac cycle T _m (k) is the duration of the current cardiac cycle calculated by the gray prediction model GM(1,1) according to the fluctuation law of the previous cardiac cycle; m is the predicted time domain length, and m>0; ξ is the error coefficient vector, which determines the weight of each error component, ξ=[ξ ₁ ξ ₂ L ξ _n ], 0<ξ _n <1, n=1, 2, 3,..., m ;

The specific implementation of the gray prediction model GM(1,1) is as follows:

Let the original data of the cardiac cycle be listed as T ⁰ (k) (k=1,2,Ln), and T ¹ (k) (k=1,2,Ln) is the accumulated sum of T ⁰ (k), that is $T^1\left(k\right)=\underset{j=1}{\overset{k}{\mathrm{\&Sigma;}}}{T}^0\left(j\right);$

The differential equation of the gray model is T ⁽⁰⁾ (k)+aT ⁽¹⁾ (k)=b k=1,2,Ln; n>0; let $Y=\left[\begin{array}{c}{T}^0\left(2\right)\\ T^0\left(3\right)\\ m\\ T^0\left(\mathrm{no}\right)\end{array}\right],$ And the parameter to be identified is θ=[a,b] ^T ,

make $B=\left[\begin{array}{cc}-(T^0\left(1\right)+T^0\left(2\right))/2& 1\\ -(T^0\left(2\right)+T^0\left(3\right))/2& 1\\ m& \\ -(T^0(\mathrm{no}-1)+T^0\left(\mathrm{no}\right))/2& 1\end{array}\right],$ θ=(B ^T B) ^-1 B ^T Y, the prediction expression T ⁰ (k+1)=T ¹ (k+1)-T ¹ (K)=(1-exp(a))[T ⁰ (1)-b/a]exp(-ak), when k<n in the above formula, predict the restored value T ⁰ (k) of the output T original data column, when k=n, predict the output T at the next moment The predicted output value of T, when k>n, the predicted output value after k+n cardiac cycles of predicted output T.

Preferably, in the above-mentioned predictive adaptive control system for medicinal liquid bolus injection, the microcontroller in the medicinal liquid bolus injection module is connected to the predictive control algorithm module of the main control module; The stepping motor is connected to a screw rod through a coupling, and the other end of the screw rod is fixed on the side opposite to the stepping motor, and the screw rod is connected to a syringe plug; the medical syringe is fixed on The shell of the micro propeller, the tail end of the piston rod of the medical syringe is adapted to the position of the top of the syringe. The stepper motor drives the head of the syringe to move through the coupling, and the screw rod can convert the rotational motion of the coupling into the linear motion of the head of the syringe, so that the head of the syringe can push the piston rod to inject the liquid medicine. the

Preferably, in the above-mentioned predictive adaptive control system for bolus injection of medicinal liquid, the housing of the micro-propeller is also provided with a first limit switch and a second limit switch for limiting the movement of the syringe head, when the syringe When the inner piston is in contact with the top of the syringe, the first limit switch is in contact with the top of the syringe, the distance between the first limit switch and the second limit switch is less than the length of the syringe, and the second limit switch Located in the direction of movement of the piston rod and away from the side of the barrel. the

Preferably, in the above-mentioned predictive adaptive control system for bolus injection of medicinal fluid, the first limit switch and the second limit switch are connected with an alarm device. When the first limit switch is in contact with the top of the syringe, an alarm is issued to notify the medical staff that the injection of the medicinal liquid in the syringe is completed, and the medical staff can be reminded to replace the syringe in time. When the second limit switch is in contact with the top of the syringe, an alarm is given to remind the medical personnel that the piston rod in the syringe has reached the final position. the

Preferably, in the above-mentioned predictive adaptive control system for bolus injection of medicinal liquid, a pressure strain gauge is further provided between the top of the syringe and the tail end of the piston rod. The microcontroller calculates the pressure sensed by the pressure strain gauge, and controls the speed of the stepping motor, so that the thrust of the stepping motor is balanced with the back thrust of the blood pressure, and the liquid medicine can be prevented from being pushed out when the liquid medicine is not injected. The needle barrel and the blood will not flow back, and it is in a state of pressure balance. the

The speed and amount of drug pushing in the drug injection injection module are controlled by the main control module, which improves the reaction speed and real-time control ability of the drug injection injection module, and can realize the diastolic phase phase when the blood flow velocity is slow. The bolus injection of the medicinal liquid prevents the medicinal liquid from being rapidly diluted by the high-speed blood flow when it enters the blood vessels of the human body, thereby maintaining a higher concentration, which is more conducive to the medicinal liquid killing cancer cells, thus realizing scientific, rational and safe drug use, that is, with With less drug dose, the maximum drug concentration can be obtained in local organs and stay long enough without reflux, reducing damage to healthy tissues. Not only has it changed many clinical problems caused by non-standard, high-speed, and large-dose injections that cannot be ignored by manual bolus injections in the current transvascular interventional drug push, it has also greatly reduced the economic burden on the country and patients, and reduced the number of medical staff. stamina and energy. the

Compared with prior art, beneficial effect of the present invention:

1. The drug pushing speed and the drug pushing amount in the Chinese medicine liquid injection module of the present invention are controlled by the main control module, which improves the response speed and real-time control ability of the medicinal liquid injection module, and can realize that the blood flow speed is slow. The injection of the medicinal solution during the diastolic phase prevents the medicinal solution from being rapidly diluted by the high-speed blood flow when it enters the blood vessels of the human body, thereby maintaining a higher concentration, which is more conducive to the medicinal solution to kill cancer cells, thus achieving a scientific, reasonable, and Safe drug use, that is, to obtain the maximum drug concentration in local organs with a small drug dose and stay for a long enough time without reflux, so as to reduce damage to healthy tissues. the

2. The present invention uses microcontrollers, micro-propellers, and medical syringes to form a drug injection module, which not only changes the non-standard, high-speed, and large-dose drug injection that cannot be ignored in the current transvascular interventional drug injection. Many clinical problems have also greatly reduced the economic burden on the country and patients, and reduced the physical strength and energy of medical staff. the

3. Since the present invention uses a predictive control algorithm to calculate the cardiac cycle and the characteristic parameters in the cycle, according to the calculated cardiac cycle and the characteristic parameters in the cycle, the time when the medicinal solution released in the next cardiac cycle enters the blood vessel entrance is predicted. In contrast, starting the drug injection module in advance can adjust the time of pushing the drug in real time, so as to avoid causing discomfort to the patient and affecting the treatment outcome. the

4. Since the medicine pushing time of the present invention is calculated according to the patient's blood pressure value, cardiac cycle and characteristic parameters in the cycle, the medical staff can also see the dynamic waveform of the patient's blood pressure while observing the medicine pushing situation, which is convenient for the medical staff Quickly handle emergencies. the

Description of drawings:

Fig. 1 is a schematic diagram of the principle of the predictive adaptive control system for medicinal liquid bolus injection according to the present invention. the

Fig. 2 is a schematic structural view of the traditional Chinese medicine liquid injection module of the present invention. the

Fig. 3 is a flow chart of the predictive adaptive control method for bolus injection of medicinal liquid according to the present invention. the

Marks in the figure: 1-housing, 2-stepper motor, 3-coupling, 4-first limit switch, 5-second limit switch, 6-screw, 7-syringe head, 8-pressure Strain gauge, 9-piston rod, 10-needle. the

Detailed ways

The present invention will be further described in detail below in conjunction with test examples and specific embodiments. However, it should not be understood that the scope of the above subject matter of the present invention is limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention. the

Example

A predictive adaptive control system for medicinal liquid bolus injection, as shown in Figure 1, includes a main control module and a medicinal liquid bolus injection module, the main control module includes a detection module for collecting human blood pressure values, the The detection module transmits the blood pressure value to the memory area for storage to form an array of blood pressure values. The memory area is connected to an identification module, and the identification module identifies the blood pressure value array in the memory area to obtain real-time cardiac cycle and characteristic parameters in the cycle. The identification module is also connected with a predictive control algorithm module, and the predictive control algorithm module obtains the bolus injection time of the next cardiac cycle by calculating the characteristic parameters;

The medicinal liquid bolus injection module is connected to the predictive control algorithm module of the main control module, the medicinal liquid bolus injection module includes a microcontroller and a medical syringe, and the medicinal liquid bolus injection module is connected to the predictive control algorithm module through the microcontroller , the predictive control algorithm module sends the bolus injection time of the next cardiac cycle to the microcontroller, the microcontroller is connected with a micro-propeller, the micro-propeller is connected with a medical injector, and the micro-controller controls the micro-propeller The propulsion speed, injection timing and injection stop timing of the propeller, and then control the injection speed, injection injection timing and injection stop timing of the medical syringe. the

The medicinal solution bolus injection module includes a microcontroller, a micro-propeller and a medical injector, and the microcontroller is connected to the predictive control algorithm module of the main control module; as shown in Figure 2, the micro-propeller includes a housing 1 and a stepper Motor 2, the stepper motor 2 is connected with a screw rod 6 through a coupling 3, the other end of the screw rod 6 is fixed on the side opposite to the stepper motor 2, and the screw rod 6 is connected with a syringe plug 7 The medical syringe is fixed on the shell 1 of the micro-propeller, the piston rod 9 tail end of the medical syringe is adapted to the position of the syringe top 7, and the syringe top 7 and the piston rod A pressure strain gauge 8 is also provided between the 9 tail ends; a first limit switch 4 and a second limit switch 5 for limiting the movement of the syringe top 7 are also provided in the micro-propeller housing 1, when the syringe 10 When the inner piston is in contact with the top of the syringe, the first limit switch 4 is in contact with the top of the syringe 7, the distance between the first limit switch 4 and the second limit switch 5 is less than the length of the syringe 10, and The second limit switch 5 is located in the direction of movement of the piston rod 9 and away from the side of the syringe 10;

When the main control module sends the medicine pushing time of the next cardiac cycle to the medicine injection module through the predictive control algorithm module, the microcontroller starts the stepping motor to drive the coupling to rotate, and converts the rotational movement into a linear movement through the screw rod. Drive the top of the syringe to move the piston rod, and push the medicinal liquid in the syringe into the human body; the pressure strain gauge is used to sense the resistance encountered when injecting the medicinal liquid, and send the resistance to the microcontroller, and judge by the magnitude of the thrust The resistance encountered by blood vessels when the liquid medicine enters the human body; when the top of the syringe touches the first limit switch, an alarm will be sent to inform the medical staff that the injection is completed; when the top of the syringe touches the second limit switch, an alarm will be issued to notify the medical staff of the piston rod Reach the last position. The stepper motor is in a low-speed state when the drug solution injection task starts, and when the drug solution injection is not performed, the microcontroller calculates the pressure sensed by the pressure strain gauge and controls the speed of the stepper motor so that the stepper motor The propulsion force is balanced with the blood pressure return thrust. When the liquid medicine is not injected, it can prevent the liquid medicine from being pushed out of the syringe and the blood will not flow back, and it is in a state of pressure balance. the

As shown in Figure 3, the above-mentioned predictive adaptive control system for the injection of medicinal liquid processes the signal according to the following steps:

(1) Collection of blood pressure values: real-time collection of human blood pressure values through the detection module in the main control module, and recursively stored in the memory area of the main control module to form an array of blood pressure values;

(2) Obtain relevant parameters: the identification module in the main control module obtains the real-time cardiac cycle and the characteristic parameters in the cycle by identifying the array of blood pressure values; obtains the patient's symptom parameters through physical examination;

(3) Calculating the time to push the medicine: set the time to push the medicine according to the symptom parameters; the main control module calculates the time to push the medicine in the next cardiac cycle through the predictive control algorithm according to the time to push the medicine, the real-time cardiac cycle and the characteristic parameters in the cycle ;

(4) Start the medicinal solution injection module: the main control module sends the calculated drug pushing time in the next cardiac cycle to the microcontroller in the medicinal solution injection module, and the microcontroller calculates the drug pushing time according to the drug pushing time and Ahead of t _delay , start the micro injector in the liquid medicine bolus injection module to carry out the liquid medicine bolus injection, and said t _delay is the time delay of the liquid medicine bolus injection module;

(5) Self-adaptation of bolus injection: the detection module collects new blood pressure values, and compares the newly obtained sampled blood pressure values, newly identified cardiac cycle and characteristic parameters in the cycle with the previously collected blood pressure value, cardiac cycle and the characteristic parameters in the cycle. The weighted average of the characteristic parameters is used to update the parameters of the predictive control algorithm to realize the adaptability of the predictive control bolus injection of medicinal liquid. the

The medicine pushing time is related to the patient's blood pressure value, cardiac cycle and the characteristic parameters in the cycle, and the medicine pushing time can be adjusted in real time to avoid causing discomfort to the patient and affecting the treatment outcome. the

In step (1), the detection module includes a blood pressure sensor; the recursive method means that when sampling a new blood pressure value into the blood pressure value array, a group of earliest sampled blood pressure values are discarded at the same time, so that the dimension of the blood pressure value array remains unchanged. the

In step (2), the characteristic parameters in the cardiac cycle include systolic blood pressure, diastolic blood pressure, diastolic period duration, systolic period duration, etc.; symptom parameters include tumor diameter, tumor blood supply artery, tumor distance from the preset drug injection point distance etc. the

In step (3), the medicine pushing moment refers to the starting moment of a phase in the characteristic parameters of the cardiac cycle; the medicine pushing moment t _r (k+1) in the next cardiac cycle is passed through the equation t _r (k+ 1)'=t _r (k)+T _m (k)+

ξ[(t _r (k)-t _r (k)',t _r (k-1)-t _r (k-1)',Lt _r (km)-t _r (k=m)'] ^T to Calculation, where, t _r (k) is the initial moment of this phase in the current cardiac cycle, that is, the current drug pushing moment; t _r (k)' is calculated by the predictive control algorithm based on the characteristic parameters in the last cardiac cycle T _m (k) is the duration of the current cardiac cycle calculated by the gray prediction model GM(1,1) according to the fluctuation law of the previous cardiac cycle; m is the predicted time domain length, and m>0; ξ is the error coefficient vector, which determines the weight of each error component, ξ=[ξ ₁ ξ ₂ L ξ _n ], 0<ξ _n <1, n=1, 2, 3,..., m .

The specific implementation of the gray prediction model GM(1,1) is as follows:

Let the original data of the cardiac cycle be listed as T ⁰ (k) (k=1,2,Ln), and T ¹ (k) (k=1,2,Ln) is the accumulated sum of T ⁰ (k), that is $T^1\left(k\right)=\underset{j=1}{\overset{k}{\mathrm{\&Sigma;}}}{T}^0\left(j\right);$

The differential equation of the gray model is T ⁽⁰⁾ (k)+aT ⁽¹⁾ (k)=b k=1,2,Ln; n>0; let $Y=\left[\begin{array}{c}{T}^0\left(2\right)\\ T^0\left(3\right)\\ m\\ T^0\left(\mathrm{no}\right)\end{array}\right],$ And the parameter to be identified is θ=[a,b] ^T ,

make $B=\left[\begin{array}{cc}-(T^0\left(1\right)+T^0\left(2\right))/2& 1\\ -(T^0\left(2\right)+T^0\left(3\right))/2& 1\\ m& \\ -(T^0(\mathrm{no}-1)+T^0\left(\mathrm{no}\right))/2& 1\end{array}\right],$ θ=(B ^T B) ^-1 B ^T Y, get the prediction expression T ⁰ (k+1)=T ¹ (k+1)-T ¹ (K)=(1-exp(a))[T ⁰ (1)-b/a]exp(-ak), when k<n in the above formula, predict the restored value T ⁰ (k) of the output T original data column, when k=n, predict the output T at the next moment The predicted output value of T, when k>n, the predicted output value after k+n cardiac cycles of predicted output T.

Claims (5)

1. a prediction adaptive control system of injecting for medicinal liquid, comprises that main control module and medicinal liquid inject module, it is characterized in that, described main control module comprises the detection module for gathering human blood-pressure value, and described detection module comprises pressure transducer; Described detection module is passed to memory field by this pressure value and stores, form pressure value array, described memory field is connected with identification module, the pressure value array of described identification module identification memory field obtains real-time cardiac cycle and the characteristic parameter in the cycle, described identification module is also connected with predictive control algorithm module, and described predictive control algorithm module is by calculating injecting the moment of next cardiac cycle to characteristic parameter;

The predictive control algorithm module of main control module injected module and is connected in by described medicinal liquid, described medicinal liquid is injected module and is comprised microcontroller and injector for medical purpose, described medicinal liquid is injected module and is connected in predictive control algorithm module by microcontroller, the moment of injecting of next cardiac cycle is sent to microcontroller by described predictive control algorithm module, described microcontroller is connected with micro-thruster, described micro-thruster is connected with injector for medical purpose, the fltting speed of described microprocessor controls micro-thruster, inject the moment and inject and stop the moment, and then the injection speed of control injector for medical purpose, inject the moment and inject and stop the moment,

Described system is according to steps of processing signal:

(1) gather pressure value: by the detection module Real-time Collection human blood-pressure value in main control module, the memory field being stored in main control module in the mode of recursion forms pressure value array;

(2) obtain relevant parameter: the identification module in main control module obtains real-time cardiac cycle and the characteristic parameter in the cycle by identification pressure value array; Obtain patient's symptom parameter by health check-up;

(3) calculate and push away the medicine moment: setting pushes away the medicine moment according to symptom parameter; Main control module, according to pushing away medicine moment, real-time cardiac cycle and the characteristic parameter in the cycle, calculates pushing away the medicine moment in next cardiac cycle by predictive control algorithm;

(4) start medicinal liquid and inject module: main control module is sent to the medicine moment that pushes away in the next cardiac cycle calculating medicinal liquid and injects the microcontroller in module, described microcontroller pushes away medicine moment t in advance according to this _delay, start medicinal liquid and inject in module micro-injector and carry out medicinal liquid and inject, described t _delayfor medicinal liquid is injected the time delay of module;

(5) self adaptation of injecting: the pressure value that detection module collection is new, the cardiac cycle by the sampling pressure value newly obtaining, newly identifying and the characteristic parameter in the cycle and the pressure value, cardiac cycle and the characteristic parameter weighted average in the cycle that gather before, the parameter of upgrading predictive control algorithm, realizes the adaptivity that medicinal liquid PREDICTIVE CONTROL is injected;

In step (1), the mode of recursion refers to when the new pressure value of sampling enters pressure value array, loses one group of sampling pressure value the earliest simultaneously, and the dimension of pressure value array is remained unchanged;

In step (2), the characteristic parameter in cardiac cycle comprises systolic pressure, diastolic pressure, diastole duration, systolic period duration; Symptom parameter comprises that diameter of tumor, supply artery of the tumor, tumor inject a distance apart from default medicinal liquid;

In step (3), the initial time of phase while pushing away in the characteristic parameter that the medicine moment refers to cardiac cycle; In next cardiac cycle, push away medicine moment t _r(k+1) by equation t _r(k+1) '=t _r(k)+T _m(k)+

ξ [(t _r(k)-t _r(k) ', t _r(k-1)-t _r(k-1) ', Lt _r(k-m)-t _r(k=m) '] ^tcalculate, wherein, t _r(k) be in current cardiac cycle this time phase initial time, i.e. current pushing away the medicine moment; t _r(k) ' in the current cardiac cycle that draws according to the calculation of characteristic parameters in last cardiac cycle for predictive control algorithm this time phase initial time; T _m(k) be Grey models GM (1,1) according to before the duration of the current cardiac cycle that calculates of cardiac cycle fluctuation pattern; Length of field when m is prediction, and m>0; ξ is error coefficient vector, determines the weight of each error component, ξ=[ξ ₁ξ ₂l ξ _n], 0 < ξ _n< 1, n=1,2,3 ..., m;

Being implemented as follows of Grey models GM (1,1):

If cardiac cycle initial data is classified T as ⁰(k) (k=1,2, Ln), T ¹(k) (k=1,2, Ln) is T ⁰(k) one-accumulate and, $T^1\left(k\right)=\underset{j=1}{\overset{k}{\mathrm{\&Sigma;}}}{T}^0\left(j\right);$

The gray level model differential equation is T ⁽⁰⁾(k)+aT ⁽¹⁾(k)=b k=1,2, Ln; N>0; Order $Y=\left[\begin{array}{c}{T}^0\left(2\right)\\ T^0\left(3\right)\\ M\\ T^0\left(n\right)\end{array}\right],$ And parameter to be identified is θ=[a, b] ^t,

Order $B=\left[\begin{array}{cc}-(T^0\left(1\right)+T^0\left(2\right))/2& 1\\ -(T^0\left(2\right)+T^0\left(3\right))/2& 1\\ M& \\ -(T^0(n-1)+T^0\left(n\right))/2& 1\end{array}\right],$ θ=(B ^tb) ^-1b ^ty, obtains prediction expression T ⁰(k+1)=T ¹(k+1)-T ¹(K)=(1-exp (a)) [T ⁰(1)-b/a] exp (ak), in above formula, in the time of k<n, the reducing value T of T initial data row is exported in prediction ⁰(k), in the time of k=n, the prediction output valve in prediction output next moment of T, in the time of k>n, the prediction output valve after k+n the cardiac cycle of prediction output T.

2. prediction adaptive control system of injecting for medicinal liquid according to claim 1, is characterized in that, the predictive control algorithm module of main control module injected microcontroller in module and be connected in by described medicinal liquid; Described micro-thruster comprises shell and motor, and described motor is connected with screw mandrel by shaft coupling, and the described screw mandrel other end is fixed on a side relative with motor, and described screw mandrel is connected with syringe top; Described injector for medical purpose is fixed on the shell of described micro-thruster, and the position of the piston rod tail end of described injector for medical purpose and described syringe top is suitable.

3. prediction adaptive control system of injecting for medicinal liquid according to claim 2, it is characterized in that, on described micro-thruster shell, be also provided with for limiting mobile the first limit switch and the second limit switch in syringe top, in the time that syringe inner carrier contacts with syringe top, described the first limit switch contacts with syringe top, distance between described the first limit switch and the second limit switch is less than syringe length, and the second limit switch is positioned at the direction of motion of piston rod the side away from syringe.

4. prediction adaptive control system of injecting for medicinal liquid according to claim 3, is characterized in that, described the first limit switch and the second limit switch are connected with alarm device.

5. prediction adaptive control system of injecting for medicinal liquid according to claim 4, is characterized in that, between described syringe top and described piston rod tail end, is also provided with pressure-strain sheet.