Three-dimensional shape reconstruction method of magnetic continuum interventional instrument

Abstract

The invention discloses a three-dimensional shape reconstruction method for a magnetic continuum interventional instrument, and belongs to the technical field of magnetic interventional medical instruments, and the method comprises the following steps: S1, detecting a magnetic field of the magnetic interventional instrument through an external sensor; s2, establishing a corresponding relation model of the magnetic field of the guide wire magnetic section and the form thereof; and S3, fusing the intraoperative image and the magnetic field information to carry out three-dimensional form reconstruction. According to the three-dimensional shape reconstruction method of the magnetic continuum interventional instrument, the shape and the position of the interventional instrument in the body of a patient can be obtained without additionally arranging an additional sensor on the interventional instrument or changing the mechanical property of the interventional instrument, and the precision is high.

Description

Technical Field

[0001] The invention relates to the technical field of magnetic interventional medical devices, and in particular to a three-dimensional morphology reconstruction method of a magnetic continuum interventional device. Background Art

[0002] At present, stroke has become the second leading cause of death and disability in the world, and the number of stroke patients is still increasing. Among them, acute ischemic stroke (cerebral infarction) and intracranial hemorrhage caused by aneurysm rupture are the main causes of stroke. Neurointerventional surgery is the main way to treat the above diseases. Doctors observe the digital silhouette angiography (DSA) images under X-rays, use guidewires to assist in delivering catheters, microcatheters and other instruments from the patient's radial artery or femoral artery opening to the narrow intracranial blood vessels or aneurysms, establish treatment channels, and perform thrombus dissolution, extraction or aneurysm embolization treatment by infusing thrombolytic agents, releasing thrombectomy stents, placing spring coils, etc. The timeliness of stroke treatment and the complexity of cerebral blood vessels place extremely high demands on the ability of equipment to be in place. The successful and efficient establishment of interventional treatment pathways is a key issue in improving the survival rate of stroke patients and reducing the disability rate.

[0003] In minimally invasive interventional surgery, interventional devices work in narrow and closed human blood vessels. The size of interventional devices is strictly restricted, and the means of remote control are relatively limited. Low-frequency quasi-static magnetic fields are harmless to the human body and have good remote field control capabilities. They are suitable for precise control of interventional access devices. Unlike ordinary interventional access devices, the distal end of magnetic interventional access devices is made of permanently magnetized magnetic materials. Under the action of an external magnetic field, the entire or part of the magnetic interventional device can be deflected or moved as needed to achieve controllable head pointing or position movement, thereby reducing the difficulty of superselective operations on vascular branches and improving the success rate and efficiency of the device. This type of magnetic navigation interventional system contains two basic elements: a magnetic interventional device and a magnetic field control device.

[0004] For magnetic interventional devices, if you want to accurately control the shape of the magnetic interventional device at the next moment through the magnetic field control device, you must obtain the current position and shape of the magnetic interventional device in advance. At the same time, for doctors performing operations, since most of the current cases use two-dimensional perspective images to assist interventional surgery, the spatial pointing of the interventional device is unclear. Therefore, doctors are required to conduct multiple attempts to estimate the specific pointing and shape of the interventional device based on the specific feel, which greatly increases the risk of blood vessel rupture and prolongs the operation time. Therefore, being able to intuitively obtain the shape and position of an entire interventional device can improve the safety and efficiency of the operation, and also provide the necessary conditions for the navigation of the interventional device.

[0005] In the prior art, the following technical problems generally exist:

[0006] 1. The interventional device is located in the patient's body. The method based on ordinary fluoroscopic images cannot obtain the spatial position of the interventional device in the patient's body, and cannot establish the specific shape of the device.

[0007] 2. If active sensors are added to interventional instruments for morphological reconstruction, turning passive instruments into active instruments, the mechanical properties of the instruments will be greatly changed, and there will be problems with sensor wires, which will interfere with surgical operations. Summary of the invention

[0008] The purpose of the present invention is to provide a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device, which does not require the installation of additional sensors on the interventional device, nor does it require changing the mechanical properties of the interventional device, and can obtain the morphology and position of the interventional device in the patient's body with high accuracy.

[0009] To achieve the above object, the present invention provides a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device, comprising the following steps:

[0010] S1. Detecting the magnetic field of the magnetic interventional device through an external sensor;

[0011] S2, establishing a corresponding relationship model between the magnetic field of the guide wire magnetic segment and its morphology;

[0012] S3. Intraoperative imaging and magnetic field information are integrated to perform three-dimensional morphological reconstruction.

[0013] Preferably, in S1, the magnetic interventional instrument is a flexible interventional instrument having a flexible magnetic segment and allowing detection by an external magnetic field and a fluoroscopic imaging device.

[0014] Preferably, in S2, the magnetic segments of the guide wire are equivalent to n segments of cylindrical magnets, and the center of the cylindrical magnet is taken as the origin. In the magnet cylindrical coordinate system, for a point x(x s ,y s ,z s ), calculate the magnetic field at point x according to the method of finding the magnetic field of a cylindrical magnet;

[0015] The finite difference of the guide wire magnetic segment converts the morphology estimation of the entire magnetic segment into the estimation of the posture parameters of the n-segment cylindrical magnet. The parameters that need to be estimated for the j-th cylindrical magnet in the spherical coordinate system are (x j ,y j , z j ,θ j ,φ j ), the expression of the constructed magnetic field-morphology model is:

[0016]

[0017] Among them, x j ,y j, z j is the spatial position coordinate of the center of the cylindrical magnet, θ j ,φ j are the pitch and roll angles, The sensor measures the magnetic field, B k Calculate the magnetic field for the model.

[0018] Preferably, in S3, the two-dimensional morphology of the magnetic segment of the guide wire needs to be extracted first, and then the magnetic information is fused to reconstruct the three-dimensional morphology.

[0019] Preferably, when extracting the two-dimensional morphology of the magnetic segment of the guide wire, the pixels of the magnetic segment of the guide wire in the image are detected and segmented, and finally the two-dimensional morphological skeleton of the magnetic segment is obtained.

[0020] Preferably, when fusing magnetic information for 3D morphological reconstruction, only the parameters (z'1,θ j )(j=1,2,…,n) is estimated to obtain the three-dimensional shape of the magnetic segment of the guide wire; first, the guide wire length is taken as the target pair θ j Make a preliminary estimate.

[0021] Then, the error between the measured value of the magnetic field at the sensor network and the predicted value of the magnetic field-morphology model is used as the cost function to preliminarily estimate z'1.

[0022] The estimated θ j and z'1 as initial values ​​for the parameter θ j and z'1 to make a more refined estimate of the magnetic field measurement at the sensor network The predicted values ​​of the magnetic field-morphology model B(z'1,θ1,…,θ n ) error E B As the cost function, the parameter estimation is calculated as:

[0023]

[0024] Preferably, let L th ,θ th are the length change threshold and the angle change threshold respectively, then the constraint is expressed as:

[0025]

[0026] G jθ =|θ j -θ j-1 |≤θ th ;

[0027] Among them, G L represents the constraint of small change in total length, G jθIndicates that the j-th angle change is a constraint with a limit, and the final function constructed using Lagrange multipliers is:

[0028] L(z′1,θ1,…,θ n )=E B +λ1G L +λ j G jθ (j=2,3,…,n);

[0029] According to the above function, an optimization algorithm is used to further accurately estimate the target parameters to obtain the final shape of the guidewire.

[0030] Therefore, the beneficial effects of the three-dimensional morphology reconstruction method of the magnetic continuum interventional device of the present invention are as follows:

[0031] (1) The shape and position of the interventional device in the patient's body can be obtained without adding additional sensors to the interventional device or changing the mechanical properties of the interventional device.

[0032] (2) The doctor can obtain the shape and position of the interventional device during the operation without performing additional surgical operations, which is conducive to the safe and efficient conduct of the operation.

[0033] (3) Compared with solutions such as embedded optical fibers and electromagnetic sensors, this method does not require the addition of additional signal wires and does not affect the routine surgical process.

[0034] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a step diagram of an embodiment of a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to the present invention;

[0036] Figure 2 Schematic diagram of the structure of a flexible magnetic tip guidewire in an embodiment of a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device of the present invention, wherein (a) is a structural diagram of the guidewire composition, and (b) is a schematic diagram of the outer layer wrapping material of the guidewire;

[0037] Figure 3 This is a magnetic sensor network structure diagram used in a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to an embodiment of the present invention;

[0038] Figure 4 The invention discloses a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device, in which a magnetic segment is finitely differentiated into n segments of a cylindrical magnet diagram. DETAILED DESCRIPTION

[0039] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0040] Embodiment 1

[0041] like Figure 1 As shown, the present invention provides a method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device, comprising the following steps:

[0042] S1. Detect the magnetic field of the magnetic interventional device through an external sensor array.

[0043] like Figure 2 As shown in (a), the magnetic interventional device used in this embodiment is a flexible guidewire with a flexible magnetic segment that allows detection by an external magnetic field and fluoroscopic imaging equipment. The inner core of the inner layer of the guidewire uses a nickel-titanium alloy wire with an optimized taper. The guidewire is prepared by a heated extrusion method and magnetized by a high-voltage pulse circuit. At the same time, in order to reduce friction and improve biocompatibility, a hydrophilic coating is covered on the outermost layer of the guidewire. In order to make the magnetic tip clearly visible in the fluoroscopic image, a radiopaque platinum-iridium coil is wrapped around the magnetic tip. As shown Figure 2 As shown in (b), the outer layer of the guidewire adopts a continuous structure in which magnetic nanoparticles are dispersed in a polyurethane matrix, and the isotropic design enhances the magnetic response capability. The material of the magnetic nanoparticles is NdFeB particles coated with silicon dioxide, which can effectively improve biocompatibility.

[0044] Tests show that the guidewire performs well in terms of softness, torsion control performance and resistance to tensile fracture, meeting the requirements of clinical use. It also has good magnetic response capabilities and is suitable for flexible operations in small blood vessels.

[0045] like Figure 3 As shown, in this embodiment, a 4×4 arranged magnetic sensor network is designed to obtain the magnetic field information of the above-mentioned n-segment cylindrical magnet. It is also possible to construct a magnetic sensor network with other structures or use other magnetic field sensing devices to sense the magnetic segments of the guide wire according to actual needs.

[0046] S2. Establish a corresponding relationship model between the magnetic field of the guide wire magnetic segment and its shape.

[0047] The guidewire tip is a flexible continuum with a complex shape. Compared with a hard permanent magnet, its magnetic field distribution changes with the shape. Therefore, the shape of the magnetic segment can be estimated by the magnetic field, so a corresponding relationship model between the magnetic field of the guidewire magnetic segment and its shape is established. Since the magnetism of any small segment of the guidewire in any shape is only related to the magnetization intensity of this segment and has nothing to do with other adjacent guidewire segments, the guidewire magnetic segment can be regarded as composed of multiple axially magnetized cylindrical magnets.

[0048] In order to model the spatial magnetic field generated by the magnetic segment at the end of the guidewire, the magnetic segment of the guidewire is divided into n cylindrical magnets by finite difference, such as Figure 4In this embodiment, for a single axially magnetized cylindrical magnet with a height of 2l and a radius of r, its magnetic field is equivalent to an energized solenoid, and the equivalent magnetic moment m = 2lnIπa 2 , taking the center of the cylindrical magnet as the origin, in the magnet cylindrical coordinate system, for a point x(x s ,y s ,z s ), according to the Biot-Savart law, the magnetic field generated by the surface strip of the solenoid is integrated to calculate the magnetic field at point x. (Among them, the modeling of n-segment cylindrical magnets can also be equivalent to dipoles or magnetic rings, or the magnetic field of each segment of the cylindrical magnet can be described and modeled using a neural network method.)

[0049] The finite difference of the guide wire magnetic segment converts the morphology estimation of the entire magnetic segment into the estimation of the posture parameters of the n-segment cylindrical magnet. The parameters that need to be estimated for the j-th cylindrical magnet in the spherical coordinate system are (x j ,y j , z j ,θ j ,φ j ), the expression of the constructed magnetic field-morphology model is:

[0050]

[0051] Among them, x j ,y j , z j is the spatial position coordinate of the center of the cylindrical magnet, θ j ,φ j are the pitch and roll angles, The sensor measures the magnetic field, B k Calculate the magnetic field for the model.

[0052] The three-dimensional posture of the guidewire is reconstructed by combining the magnetic sensor network constructed in S1 with the magnetic field-morphology model.

[0053] S3. Intraoperative imaging and magnetic field information are integrated to perform three-dimensional morphological reconstruction.

[0054] First, the two-dimensional morphology of the magnetic segment of the guidewire needs to be extracted, and then the magnetic information is fused to reconstruct the three-dimensional morphology.

[0055] In the process of estimating the magnetic segment of the guidewire, the spatial magnetic field information can generally reflect the three-dimensional shape of the magnetic segment more accurately. However, since the optimization estimation using only magnetic field information requires more parameters to be estimated and it is difficult to determine the initial value, it has a great impact on the estimation efficiency and accuracy. Using only DSA image information will face the problem of ambiguity in the three-dimensional shape of the guidewire. Therefore, the image information is fused with the magnetic field information, and the inherent permanent magnetic properties of the magnetic guidewire are utilized. Therefore, there is no need to perform additional processing on the guidewire during estimation, avoiding the addition of additional sensors to the guidewire to make the guidewire active from passive, resulting in the guidewire carrying a signal wire and causing a blind end closure problem, as well as the loss of guidewire performance due to the addition of sensors.

[0056] Therefore, when estimating the three-dimensional morphology of the guidewire, the two-dimensional morphology of the guidewire in the two-dimensional fluorescent fluoroscopy image can be extracted to form morphological prior information and provided to the morphological estimation algorithm through the magnetic field, thereby improving the calculation efficiency and the accuracy of the estimation results. In addition, DSA imaging is an imaging modality that must be used during surgery and does not add any additional burden to the equipment and operation.

[0057] When extracting the two-dimensional shape of the guide wire magnetic segment, the first step is to detect the target and segment the pixels of the guide wire magnetic segment, then extract the key points of the starting point and the ending point of the guide wire magnetic segment, and finally extract the skeleton of the pixel part of the guide wire magnetic segment.

[0058] In this embodiment, when performing target detection on the head end of the guide wire magnetic segment, the RTMDet target detector is used. For the task of segmenting the guide wire magnetic segment, the DeepLabv3+ model is used to perform segmentation within the range of target detection. On the basis of completing the target detection and pixel segmentation of the guide wire magnetic segment, in order to accurately obtain the magnetic segment part, it is necessary to identify the key points of the starting point and the end point of the guide wire magnetic segment. In this embodiment, RTMPose is used as a key point recognition model to identify the key points. When performing skeleton extraction on the guide wire magnetic segment pixels, the Zhang-Suen algorithm is used. After the skeleton pixels are extracted, the pixel closest to the starting point of the magnetic segment in the skeleton is found as the starting point of the magnetic segment skeleton, and the end point pixels are added to the skeleton to form the final guide wire magnetic segment skeleton.

[0059] When fusion magnetic information is used for 3D morphological reconstruction, the relevant posture parameters can be directly obtained because the 2D morphological information of the guidewire can be obtained based on the perspective image. Then, it is set as the bottom height of the j-th cylindrical magnet. Since the bottom height of the j+1-th cylindrical magnet can be deduced from the bottom height of the j-th cylindrical magnet, only the parameters (z'1,θ j )(j=1,2,…,n) is estimated to obtain the three-dimensional shape of the guide wire magnetic segment.

[0060] First, the guide wire length is taken as the target j A preliminary estimate is made, and then the error between the measured value of the magnetic field at the sensor network and the predicted value of the magnetic field-morphology model is used as the cost function to preliminarily estimate z'1. Finally, the estimated θ j and z'1 as initial values ​​for the parameter θ j and z'1 to make a more refined estimate of the magnetic field measurement at the sensor network The predicted values ​​of the magnetic field-morphology model B(z'1,θ1,…,θ n ) error E B As the cost function, the parameter estimation is calculated as:

[0061]

[0062] Also, since the magnetic segment has a small change in length during operation and the overall bending degree is not too large, the optimization parameters need to satisfy two nonlinear constraints. Let L th ,θ th are the length change threshold and the angle change threshold respectively, then the constraint is expressed as:

[0063]

[0064] G jθ =|θ j -θ j-1 |≤θ th ;

[0065] Among them, G L represents the constraint of small change in total length, G jθ Indicates that the j-th angle change has a certain limit constraint, and the final function constructed using Lagrange multipliers is:

[0066] L(z′1,θ1,…,θ n )=E B +λ1G L +λ j G jθ (j=2,3,…,n);

[0067] According to the above function, an optimization algorithm is used to further accurately estimate the target parameters to obtain the final shape of the guidewire.

[0068] Therefore, the present invention adopts the above-mentioned three-dimensional morphology reconstruction method of the magnetic continuum interventional device, which does not require the installation of additional sensors on the interventional device, nor does it require the change of the mechanical properties of the interventional device. It can obtain the morphology and position of the interventional device in the patient's body with high accuracy.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device, characterized in that: The following steps are involved: S1. Detecting the magnetic field of the magnetic interventional device through an external sensor; S2, establishing a corresponding relationship model between the magnetic field of the guide wire magnetic segment and its morphology; S3. Intraoperative imaging and magnetic field information are integrated to perform three-dimensional morphological reconstruction.

2. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 1, characterized in that: In S1, the magnetic interventional device is a flexible interventional device having a flexible magnetic segment and allowing detection by an external magnetic field and a fluoroscopic imaging device.

3. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 1, characterized in that: In S2, the magnetic segments of the guide wire are equivalent to n segments of cylindrical magnets. The center of the cylindrical magnet is taken as the origin. In the magnet cylindrical coordinate system, for a point x(x s ,y s ,z s ), calculate the magnetic field at point x according to the method of finding the magnetic field of a cylindrical magnet; The morphology estimation of the entire magnetic segment is converted into the estimation of the posture parameters of the n-segment cylindrical magnet. The parameters that need to be estimated for the j-th cylindrical magnet in the spherical coordinate system are (x j ,y j , z j ,θ j ,φ j ), the expression of the constructed magnetic field-morphology model is: Among them, x j ,y j , z j is the spatial position coordinate of the center of the cylindrical magnet, θ j ,φ j is the pitch angle and roll angle, B~ k The sensor measures the magnetic field, B k Calculate the magnetic field for the model.

4. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 3, characterized in that: In S3, the two-dimensional morphology of the magnetic segment of the guidewire needs to be extracted first, and then the magnetic information is fused to reconstruct the three-dimensional morphology.

5. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 4, characterized in that: When extracting the two-dimensional morphology of the magnetic segment of the guide wire, the pixels of the magnetic segment of the guide wire in the image are detected and segmented, and finally the two-dimensional morphological skeleton of the magnetic segment is obtained.

6. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 4, characterized in that: When integrating magnetic information for 3D morphological reconstruction, only the parameters (z'1,θ j )(j=1,2,…,n) is estimated to obtain the three-dimensional shape of the magnetic segment of the guide wire; first, the guide wire length is taken as the target pair θ j Make a preliminary estimate, Then, the error between the measured value of the magnetic field at the sensor network and the predicted value of the magnetic field-morphology model is used as the cost function to preliminarily estimate z'1. The estimated θ j and z'1 as initial values ​​for the parameter θ j and z'1 to make a more refined estimate of the magnetic field measurement at the sensor network The predicted values ​​of the magnetic field-morphology model B(z'1,θ1,…,θ n ) error E B As the cost function, the parameter estimation is calculated as:

7. The method for reconstructing the three-dimensional morphology of a magnetic continuum interventional device according to claim 6, characterized in that: Let L th ,θ th are the length change threshold and the angle change threshold respectively, then the constraint is expressed as: G jθ =|θ j -θ j-1 |≤θ th ; Among them, G L represents the constraint of small change in total length, G jθ Indicates that the j-th angle change is a constraint with a limit, and the final function constructed using Lagrange multipliers is: L(z'1,θ1,…,θ n )=E B +λ1G L +λ j G jθ (j=2,3,…,n); According to the above function, an optimization algorithm is used to further accurately estimate the target parameters to obtain the final shape of the guidewire.

Images