Interventional operation continuum catheter tail end stress estimation method and force feedback system

Abstract

The invention discloses an interventional operation continuum catheter tail end stress estimation method and a force feedback system, and aims to solve the problem that catheter tail end force feedback lags behind or is low in accuracy in the interventional operation process. The continuum catheter tail end stress estimation method comprises the following steps: applying a beam theory to elastic mechanical modeling of a catheter, and performing force / moment balance analysis on a model under the condition of considering tail end external force so as to establish an elastic mechanical model of the catheter; a plurality of data sets of external force, internal force and joint configuration are obtained through an elastic mechanical model to form a training set and a test set of a BP neural network model, the BP neural network model is trained to obtain constant parameters of a nonlinear mapping function with the internal force and instantaneous joint configuration as variables and the external force as a dependent variable, and estimation of the external force borne by the catheter is achieved. Simulation test results show that the continuum catheter tail end stress estimation method has good reliability and feasibility.

Description

Technical Field

[0001] The present invention relates to the technical field of interventional surgery control, and more specifically, to a method for estimating the force applied to the distal end of a continuous catheter in an interventional surgery and a force feedback system. Background Art

[0002] With the development of medical technology, interventional treatment and minimally invasive surgery are becoming more and more common. These surgical methods use special catheters, guidewires and other instruments to reach the location of vascular lesions for diagnosis and treatment, which has the advantages of less trauma, faster recovery and fewer complications. However, these operations require extremely high precision and stability, and force feedback at the end of the catheter becomes an important factor in ensuring the safety and effectiveness of the operation.

[0003] The operating environment of the catheter in the human body is complex. Faced with different vascular pathways, tissue structures, and lesion sites, doctors need to accurately control the movement of the catheter and the force applied. Lack of effective force feedback may lead to the following problems: 1. Excessive force may damage blood vessels or surrounding tissues, leading to serious complications such as bleeding and perforation. 2. Lack of accurate force feedback may cause the catheter to fail to accurately reach the target location, affecting the surgical effect.

[0004] The force feedback system can provide real-time force information to help doctors control the catheter more accurately during operation, avoid tissue damage, and improve the safety and effectiveness of the operation. The current catheter tip force feedback is estimated based on the mechanical end sensor data combined with the method, which has problems such as sensor noise, insufficient accuracy, and response lag (especially during rapid operation, the feedback lag problem affects the overall effect). Therefore, it is necessary to develop a new catheter tip force estimation method to overcome these problems. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention proposes a method for estimating the force on the end of a continuum catheter during interventional surgery and a force feedback system, aiming to solve the problems of lack of force feedback and accuracy at the end of the catheter during interventional surgery.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for estimating the force on the end of a continuous catheter in an interventional surgery, comprising:

[0007] Step S1, establishing a mechanical model of the continuum catheter: assuming that a known external force is applied to the end portion of the continuum catheter, a mechanical model of the continuum catheter is established based on the Correrat beam theory to represent the joint configuration, internal force and internal torque relationship of the end portion of the continuum catheter;

[0008] Step S2, solving the mechanical model, that is, solving ordinary differential equations under satisfying boundary conditions;

[0009] Step S3, calculating the elastic potential energy of the continuum catheter through the mechanical model, and obtaining the elastic mechanical model of the coupling of the internal force, instantaneous joint configuration and catheter external force according to the principle of virtual work, the expression is:

[0010]

[0011] In the formula, is the external force applied to the end of the continuum catheter, is the gradient of the elastic potential energy with respect to the perturbation in the configuration space, which is related to the material properties of the continuum catheter and the instantaneous joint configuration related, and is the velocity Jacobian matrix, which is included in the built-in Parameters; is the internal force of the continuum catheter;

[0012] Step S4: Use an internal force and instantaneous joint configuration Nonlinear mapping function as independent variable To describe the external force , expressed as: ; Constructing a BP neural network model; using joint configuration data and internal force as input features of the BP neural network model, and the corresponding end external force as output features;

[0013] Step S5, inputting the randomly generated external force and internal force into the elastic mechanics model of step S3, calculating and obtaining the instantaneous joint configuration of the catheter corresponding to each input value, and obtaining the training set and the test set of the BP neural network model; training the BP neural network model to obtain the constant parameters of the nonlinear mapping function;

[0014] Step S6: using the nonlinear mapping function constant parameters obtained by the trained BP neural network model to estimate the external force on the end of the continuum catheter.

[0015] The continuum catheter end force estimation method applies the Correrat beam theory to the mechanical modeling of the catheter, thereby solving the catheter internal force and joint configuration parameters under known external force conditions. Combining the catheter mechanical model and the principle of virtual work, the elastic mechanical model relationship between the catheter internal force, instantaneous joint configuration and catheter external force coupling is obtained; through this relationship, multiple data sets of external force, internal force and joint configuration are obtained. The calculated data sets are composed of the training set and test set of the BP neural network model to train the BP neural network model, and finally the internal force and instantaneous joint configuration are obtained as variables, and the constant parameters of the training nonlinear mapping function are used to achieve the estimation of the external force on the output catheter.

[0016] Furthermore, in step S2, the mechanical model is solved by using a shooting method: first, the unknown variable Set a random guess value , where k represents the number of guessing cycles; then, after determining the guessed value, the formulas in the positive elastic mechanics model are integrated in turn to solve the internal force at the end of the central rod of the catheter, and then the current error vector is obtained according to the boundary function , and obtain the residual Jacobian matrix ; Set threshold , when satisfied When the current catheter end position information is output, otherwise, the unknown variables are updated again. , repeat the integral calculation in the positive elasticity model again until the vector Until the conditions are met.

[0017] Furthermore, the BP neural network model constructed in step S5 includes 8 input neurons, 2 output neurons and 2 hidden layers, each hidden layer includes 80 neurons, and the Sigmoid function is selected as the activation function.

[0018] Furthermore, in step S5, each input feature of the BP neural network model is normalized to the minimum and maximum, which is conducive to obtaining a faster convergence speed and training effect.

[0019] Furthermore, in step S7, the trained BP neural network model is used to estimate the external force sequences that vary in a sinusoidal curve and a triangular shape.

[0020] In a second aspect, the present invention provides an interventional surgery continuum catheter tip force feedback system, comprising estimating the interventional surgery continuum catheter tip force and feeding back the estimated catheter tip force to a catheter controller; the interventional surgery continuum catheter tip force estimation method as described above is used to estimate the interventional surgery continuum catheter tip force.

[0021] Compared with the prior art, the present invention has the following technical effects:

[0022] (1) The present invention establishes a mechanical model of the catheter based on the Correrat beam theory, takes the external force into consideration in the mechanical model, and derives the mapping relationship between the internal force of the continuum catheter and the catheter shape (characterized by the joint configuration). Then, combined with the principle of virtual work, an elastic mechanical model of the coupling of internal force, instantaneous joint configuration and external force of the catheter is obtained. The simulation results in MATLAB are numerically close to the position error under the load positioning experiment, which verifies the reliability of the mechanical modeling.

[0023] (2) The present invention estimates the external force on the end of the continuum catheter through the BP neural network model, which solves the difficulty of solving the nonlinear problem and effectively estimates the end force feedback.

[0024] (3) The present invention uses the trained BP neural network model to estimate the external force sequences with sinusoidal and triangular changes, respectively, and obtains good estimation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The figure is a flow chart of a method for estimating the force on the end of a continuum catheter according to an embodiment of the present invention.

[0026] Figure 2 Schematic diagram of beam force in one embodiment of the present invention.

[0027] Figure 3 It is a schematic diagram of a continuous tube being bent by an external force in one embodiment of the present invention.

[0028] Figure 4 Schematic diagram of the BP neural network model architecture in one embodiment of the present invention.

[0029] Figure 5 This is a test flow chart of the BP neural network model predicting external force in one embodiment of the present invention.

[0030] Figure 6 This is a graph showing the prediction test results of a sinusoidal external force sequence in one embodiment of the present invention.

[0031] Figure 7 This is a prediction error diagram of a sinusoidal external force sequence in one embodiment of the present invention.

[0032] Figure 8 This is a diagram of the prediction test results of a triangle external force sequence in one embodiment of the present invention.

[0033] Fig. 9 This is a diagram of the prediction error of a triangular external force sequence in one embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0035] In the following detailed description, many specific details are set forth to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that well-known algorithms and models are not shown in detail to avoid obscuring the subject matter of the present invention.

[0036] In addition, the order of execution of actions, steps, etc. in the devices and methods shown in the claims, specifications and drawings can be implemented in any order as long as there is no special explicit limitation on the order and the output of the previous processing is not used in the subsequent processing.

[0037] Example 1

[0038] See also Figure 1 This embodiment provides a method for estimating the force on the end of a continuous catheter in an interventional surgery, comprising:

[0039] Step S1, establishing a mechanical model of the continuum catheter: Assuming that a known external force is applied to the end portion of the continuum catheter, a mechanical model of the continuum catheter is established based on the Correrat beam theory to represent the joint configuration, internal force and internal torque relationship of the end portion of the continuum catheter.

[0040] Taking the continuum catheter with two continuum segments as an example, the driving skeleton of the catheter is approximated as an ideal rod that can be sheared / stretched and bent / torqued. The force on a beam from point c to point s is as follows: Figure 2 As shown. Based on the existing Correrat beam theory, The mechanical equilibrium differential equation is integrated within the range to obtain the position, internal force and internal torque information of the end of the proximal continuous segment as follows:

[0041] (1)

[0042] In formula (1), P(s) represents the position of the end of the conduit corresponding to point s on the beam, n(s) represents the internal force of the beam corresponding to point s on the beam, m(s) represents the moment corresponding to point s on the beam, R(s) represents the rotation matrix corresponding to point s on the beam, and I 3×3 Represents the identity matrix.

[0043] Then, in middle,

[0044] (2)

[0045] In formula (2), d represents the length between two consecutive segments.

[0046] At the same time, the internal forces and internal moments at the end of the continuum tube must meet the following boundary conditions:

[0047] (3)

[0048] In formula (3), Indicates internal force, represents torque, Indicates external force, Represents external torque.

[0049] The above equations constitute the mechanical model of this continuum catheter, that is, when the external force and external torque are known Given two pairs of internal forces , the bending shape and terminal posture of the continuum catheter can be calculated. The bending shape and terminal posture are specifically described by the catheter joint configuration parameters. The inverse model of the mechanical model can calculate two pairs of internal forces under the given terminal posture condition .

[0050] Step S2, solving the mechanical model, that is, solving ordinary differential equations under satisfying boundary conditions.

[0051] We give an example of solving the mechanical model using the existing target shooting method as follows. First, give the unknown variable Set a random guess value , where k represents the number of guessing cycles. Then, after determining the guessed value, the formulas in the positive elastic mechanics model are integrated in turn to solve the internal force at the end of the central rod of the catheter, and then the current error vector is obtained according to the boundary function. , and obtain the residual Jacobian matrix . Set the threshold , when satisfied When , the current catheter end position information is output. Otherwise, the unknown variables are updated again. , repeat the integral calculation in the positive elasticity model again until the vector Until the conditions are met. The pseudo code of the mechanical model solution algorithm is shown in Table 1.

[0052] Table 1 Pseudo code of the algorithm for solving the mechanical model by the shooting method

[0053]

[0054] Step S3, calculating the elastic potential energy of the continuum conduit through the mechanical model, and performing force / torque balance analysis on the model according to the principle of virtual work:

[0055] See also Figure 3 , the elastic potential energy is:

[0056]

[0057] In the above formula,

[0058] external force and internal force Virtual work done:

[0059] (4)

[0060] in: (5) is the virtual displacement of the terminal space, the Jacobian matrix Map the end force to the generalized coordinate space. (6) is the virtual displacement in the joint space.

[0061] The change in elastic potential energy is expressed as:

[0062] (7)

[0063] According to the principle of virtual work, combined with (4)-(7) under equilibrium conditions:

[0064]

[0065] Due to virtual displacement Any, which can be expressed as:

[0066] (8)

[0067] In formula (8), is the external force applied to the end of the continuum catheter, is the gradient of the elastic potential energy with respect to the perturbation in the configuration space, which is related to the material properties of the continuum catheter and the instantaneous joint configuration related, and is the velocity Jacobian matrix, which is included in the built-in Parameters; is the internal force of the continuum catheter.

[0068] Step S4: When the catheter tip is subjected to external force, the internal force of the catheter will change according to the external force, and there is a nonlinear relationship between the external force, the internal force and the joint configuration that is difficult to describe. and instantaneous joint configuration Nonlinear mapping function as independent variable To describe the external force , expressed as: , and then construct a BP neural network model to obtain the nonlinear mapping function The constant parameter in .

[0069] Construct a BP neural network model; use joint configuration data and internal force as input features of the BP neural network model, and use the corresponding terminal external force as output features.

[0070] As an example, see Figure 4 ,The BP neural network model contains 8 input neurons, 2 output neurons and 2 hidden layers, each hidden layer contains 80 neurons, and the Sigmoid function is selected as the activation function.

[0071] Step S5, the randomly generated external force and internal force are input into the elastic mechanics model of step S3, the instantaneous joint configuration of the catheter corresponding to each input quantity is calculated, and the training set and test set of the BP neural network model are obtained; the BP neural network model is trained to obtain the constant parameters of the nonlinear mapping function.

[0072] As a preferred example, according to the actual working situation, the external force on the end of the continuum catheter is generally a two-dimensional vector, so we assume that the end force vector is located in the xy plane. We randomly generate 20,000 sets of end external force vectors and internal force vectors to drive the continuum catheter to present random bending angles and directions. These generated external and internal force values ​​are all within the tolerance range of the continuum catheter. Then, the generated external and internal forces are input into the elastic mechanics model, and the current joint configuration of the catheter corresponding to each input quantity is collected at the same time. 20,000 sets of end external force vectors, internal force vectors and their corresponding joint configuration data constitute the data set for BP neural network model training. In order to obtain faster convergence speed and training effect, each input feature is normalized to the minimum and maximum. 90% of the data set is used as a training set, and the rest is used as a test set. The network model is implemented in matlab2019, and the batch size is 32 and the number of iterations is 100 during training. During the training process, the learning rate gradually decreases from 0.01 according to the number of iterations. Once the performance on the test set starts to deteriorate, training will stop and the network parameters at this time will be saved as the best parameters.

[0073] The trained BP neural network model was verified. The elastic mechanics model of the continuum catheter established in the matlab2019 platform was used to test the reliability and feasibility of the network model.

[0074] In order to study the network model's estimation performance of the external force of the continuum catheter under different external force sequences, two sets of test experiments were conducted as follows: Figure 5 As shown in the figure, we first give a random driving rope tension input and input it into the elastic mechanics model. At the same time, a sequence of external forces with changing values ​​on the xy plane is applied to the end of the catheter. Due to the changing external force at the end, the end position and joint configuration of the continuum catheter will also change accordingly with the change of external force. Finally, we input the driving rope tension (i.e. internal force) and joint configuration at each moment into the trained BP neural network model to output the corresponding external force estimation sequence.

[0075] During the experiment, we recorded the prediction results of the BP neural network model for each external force sequence and compared them with the actual external force. In the simulation, the external force estimation error is used to represent the network prediction effect, which can be expressed by the following formula:

[0076] (9)

[0077] In formula (9), and are the external forces in the x-axis and y-axis directions estimated by the BP neural network model, and are given reference external force sequences respectively.

[0078] (1) Sinusoidal external force sequence test

[0079] In the first set of experiments, we applied a sinusoidal force sequence to the continuum catheter with a sequence length of 500. These external force sequences are divided into two components: the component amplitude in the x-axis direction is 0.3, and the component amplitude in the y-axis direction is also 0.4. The amplitude and change speed of these external force sequences are different, aiming to test the responsiveness of the BP neural network model to complex external force conditions. The estimated results of the external force components along the x-axis and y-axis are shown in Figure 6 As shown in Figure 7, the prediction error diagram of the sinusoidal external force sequence is described. It can be calculated that the maximum estimation error and the average estimation error of the external force are 0.035N and 0.017N respectively. Through this set of simulation test experiments, we verified the accuracy and stability of the BP neural network model in processing sinusoidal external force sequences.

[0080] (2) Triangular external force sequence test

[0081] In the second set of experiments, we applied an alternating force sequence of increasing and decreasing forces to the continuum robot, also known as a triangle sequence, with a sequence length of 500. This external force sequence has sudden changes and is intended to test the network model's ability to accurately predict external forces in specific working scenarios. The estimated results of the external force components along the x-axis and y-axis are shown in Figure 2. Figure 8 shown. Fig. 9 The prediction error diagram of the triangular external force sequence is described. The maximum estimation error and average estimation error of the external force are calculated to be 0.03 N and 0.015 N respectively. Through this set of experiments, we verified the performance of the network model in processing the external force sequence with sudden changes.

[0082] Step S6: using the nonlinear mapping function constant parameters obtained by the trained BP neural network model to estimate the external force on the end of the continuum catheter.

[0083] From the above method steps and test results, it can be seen that this embodiment applies beam theory to the elastic mechanics modeling of the catheter. By approximating the driving skeleton of the catheter to an ideal rod that can be sheared / stretched and bent / torqued, and considering the external force at the end, the model is subjected to force / torque balance analysis, thereby establishing an elastic mechanics model of the catheter. It is also proposed to use a BP neural network model to solve the problem of solving the nonlinear model, thereby obtaining parameters for estimating the magnitude of the external force on the catheter. The simulation test results show that this method has good reliability and feasibility.

[0084] Example 2

[0085] See also Figure 5 This embodiment provides an interventional surgery continuum catheter end force feedback system, including estimating the interventional surgery continuum catheter end force and feeding back the estimated catheter end force to the catheter controller; the interventional surgery continuum catheter end force estimation method as described in Example 1 is used to estimate the interventional surgery continuum catheter end force.

[0086] This interventional surgery continuum catheter end force feedback system can effectively speed up the catheter end force feedback speed during the interventional surgery and provide more accurate catheter force feedback.

[0087] The above-mentioned interventional surgery continuum catheter end force estimation method can be embodied in the form of a computer program product or a software functional unit. If the above-mentioned interventional surgery continuum catheter end force estimation method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Therefore, the essence of this technical solution or the part that contributes to the prior art or the part of this technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for an electronic system (which can be a personal computer, a server, or a network system, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0088] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0089] In summary, the present invention provides a method for estimating the force on the end of a continuous catheter in interventional surgery and a force feedback system, which aims to solve the problem of delayed or low accuracy of force feedback on the end of the catheter during interventional surgery. The method for estimating the force on the end of a continuous catheter includes the following steps: applying beam theory to the elastic mechanics modeling of the catheter, performing force / torque balance analysis on the model while considering the external force on the end, thereby establishing an elastic mechanics model of the catheter; obtaining multiple data groups of external force, internal force and joint configuration through the elastic mechanics model to form a training set and a test set of a BP neural network model, training the BP neural network model to obtain constant parameters of a nonlinear mapping function with internal force and instantaneous joint configuration as variables and external force as a dependent variable, thereby realizing the estimation of the external force on the catheter. Simulation test results show that the method for estimating the force on the end of a continuous catheter has good reliability and feasibility.

[0090] Those skilled in the art should understand that those skilled in the art can implement variations by combining the prior art and the above embodiments, which will not be described in detail here. Such variations do not affect the essential content of the present invention, and will not be described in detail here.

[0091] The above describes the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the above-mentioned specific embodiments, and the systems and structures that are not described in detail should be understood to be implemented in a common manner in the art; any technician familiar with the art can use the above-disclosed methods and technical contents to make many possible changes and modifications to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, or modify them into equivalent embodiments of equivalent changes, which does not affect the essential content of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention are still within the scope of protection of the technical solutions of the present invention.

Claims

1. A method for estimating the force on the end of a continuous catheter in interventional surgery, characterized in that: include: Step S1, establishing a mechanical model of the continuum catheter: assuming that a known external force is applied to the end portion of the continuum catheter, a mechanical model of the continuum catheter is established based on the Correrat beam theory to represent the joint configuration, internal force and internal torque relationship of the end portion of the continuum catheter; Step S2, solving the mechanical model, that is, solving ordinary differential equations under satisfying boundary conditions; Step S3, calculating the elastic potential energy of the continuum catheter through the mechanical model, and obtaining the elastic mechanical model of the coupling of the internal force, instantaneous joint configuration and catheter external force according to the principle of virtual work, the expression is: In the formula, is the external force applied to the end of the continuum catheter, is the gradient of the elastic potential energy with respect to the perturbation in the configuration space, which is related to the material properties of the continuum catheter and the instantaneous joint configuration related, and is the velocity Jacobian matrix, which is included in the built-in Parameters; is the internal force of the continuum catheter; Step S4: Use an internal force and instantaneous joint configuration Nonlinear mapping function as independent variable To describe the external force , expressed as: ; Constructing a BP neural network model; using joint configuration data and internal force as input features of the BP neural network model, and the corresponding end external force as output features; Step S5, inputting the randomly generated external force and internal force into the elastic mechanics model of step S3, calculating and obtaining the instantaneous joint configuration of the catheter corresponding to each input value, and obtaining the training set and test set of the BP neural network model; Training the BP neural network model to obtain constant parameters of the nonlinear mapping function; Step S6: using the nonlinear mapping function constant parameters obtained by the trained BP neural network model to estimate the external force on the end of the continuum catheter.

2. The method for estimating the force on the end of a continuous catheter in an interventional surgery according to claim 1, characterized in that: In step S2, the mechanical model is solved by using a shooting method: first, the unknown variable Set a random guess value , where k represents the number of guessing cycles; then, after determining the guessed value, the formulas in the positive elastic mechanics model are integrated in turn to solve the internal force at the end of the central rod of the catheter, and then the current error vector is obtained according to the boundary function , and obtain the residual Jacobian matrix ; Setting Thresholds , when satisfied When the current catheter end position information is output, otherwise, the unknown variables are updated again. , repeat the integral calculation in the positive elasticity model again until the vector Until the conditions are met.

3. The method for estimating the force on the end of a continuous catheter in an interventional surgery according to claim 1, characterized in that: The BP neural network model constructed in step S5 includes 8 input neurons, 2 output neurons and 2 hidden layers, each hidden layer includes 80 neurons, and the Sigmoid function is selected as the activation function.

4. The method for estimating the force on the end of a continuous catheter in an interventional surgery according to claim 1 or 3, characterized in that: In step S5, minimum and maximum normalization is performed on each input feature of the BP neural network model.

5. The method for estimating the force on the end of a continuous catheter in an interventional surgery according to claim 1, characterized in that: In step S7, the trained BP neural network model is used to estimate the external force sequences that are sinusoidal and triangular in shape.

6. A force feedback system for the end of a continuous catheter for interventional surgery, comprising estimating the force on the end of a continuous catheter for interventional surgery and feeding back the estimated force on the end of the catheter to a catheter controller; characterized in that: The force on the end of a continuous catheter for interventional surgery is estimated using the method for estimating the end of a continuous catheter for interventional surgery as described in any one of claims 1 to 5.

Images