A CT and SERF-based magnetic resonance imaging registration system and method

Abstract

This invention provides a CT and SERF-based magnetic resonance imaging (MRI) registration system and method, including a MRI registration module and a CT registration module. The MRI registration module includes a multi-turn circular positioning coil and an adjustable patch limiting device; the positioning coil can be fixed to a protrusion of the limiting device. The CT registration module includes a non-magnetic marker with an appearance consistent with the arc-shaped reference object of the MRI registration module. CT scans acquire the image structure of the human torso, perform segmentation and 3D reconstruction; the relative positional relationship between the human body and the magnetometer can be obtained based on the positioning coil. The arc-shaped reference objects of the MRI registration module and the CT registration module are registered using a nearest neighbor algorithm, thereby completing the conversion between the CT coordinate system and the magnetometer coordinate system. This system is unaffected by human respiration, improving registration accuracy, making MRI equipment more complete, and advancing subsequent 3D imaging applications.

Description

Technical Field

[0001] This invention belongs to the field of magnetic cardiac imaging technology, and particularly relates to a magnetic cardiac imaging registration system and method based on CT and SERF. Background Technology

[0002] Superconducting quantum interference devices (SQUIDs) for measuring cardiac magnetocardiography have been developing for nearly 50 years, but their widespread application is limited due to their high maintenance costs, high operating costs, and the need for operation in a magnetically shielded room, making them difficult to move. In recent years, Spin Exchange Relaxation Free (SERF) magnetometers have become a research hotspot. Alkali metal atom optically pumped magnetometers, currently the most sensitive magnetic field sensors, are gradually being used for biomagnetic measurements. Compared to SQUIDs, SERF magnetometers are smaller, operate normally at room temperature, are easier to move, and can be designed to meet different patient needs, thus possessing a broad research market. The development of cardiac magnetocardiography systems towards miniaturization and lower cost is driving the clinical application of cardiac magnetic field signals.

[0003] Three-dimensional magnetic resonance imaging (MRCI) of the heart requires multimodal registration of medical imaging structures to accurately locate lesions in patients. Before performing MRCI, the coordinate systems of different devices must be registered. Currently, a common MRCI registration method involves attaching markers to the surface of the chest cavity; during CT scanning, the same markers are then attached to the corresponding locations for MRCI registration. However, this method, based on skin surface markers, is limited in number and affected by respiratory activity, resulting in poor positioning accuracy. Reducing registration errors caused by respiration and marker errors, and improving registration accuracy to achieve higher source localization accuracy and repeatability, is a key challenge. Therefore, a high-precision MRCI registration method and system based on CT and SERF is urgently needed. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a CT and SERF-based magnetic resonance imaging registration system and method for multimodal magnetic resonance imaging (MRIE) in three dimensions.

[0005] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0006] A CT and SERF-based magnetic cardiac imaging registration system includes a magnetic cardiac registration module and a CT registration module. The magnetic cardiac registration module includes an arc-shaped reference object and a circular positioning coil located within the arc-shaped reference object. The CT registration module includes a CT registration reference object made of a non-magnetic material with the same shape as the arc-shaped reference object.

[0007] Furthermore, the magnetic registration module includes a high-precision current source to provide a stable, high-frequency, weak current signal to the circular positioning coil.

[0008] Furthermore, the arc-shaped reference object of the magnetic registration module is made of resin material, and has a cylindrical base and annular support column on top to fix the positioning coil.

[0009] Furthermore, there are two magnetic registration modules, located on both sides of the subject's torso, which are unaffected by respiratory movements.

[0010] Furthermore, the device includes a magnetocardiogram (MCC) detection device, which includes a magnetocardiogram (MCC) magnetometer array panel, a magnetocardiogram (MCC) bed, and a magnetocardiogram (MCC) shielding barrel. The MCC magnetometer array panel can be adjusted in height and angle, and the MCC bed is equipped with a magnetocardiogram (MCC) registration module with an arc-shaped reference object.

[0011] Furthermore, including CT detection equipment, before CT imaging, CT registration module registration reference objects are fixed on both sides of the subject's body, and information of the CT registration module registration reference objects is collected during CT imaging.

[0012] Furthermore, the arc-shaped reference object of the magnetic cardiomyography registration module and the registration reference object of the CT registration module are registered using a reference point and nearest neighbor algorithm.

[0013] This invention also provides a registration method for a CT and SERF-based magnetic resonance imaging registration system, comprising the following steps:

[0014] Step 1: The subject lies on the magnetic cardiopulmonary bypass bed, and the magnetic cardiopulmonary bypass panel is moved so that the center of the panel is aligned with the xiphoid process of the human body;

[0015] Step 2: Adjust the arc-shaped reference objects of the magnetic registration module on both sides of the magnetic bed to fit the sides of the human body. The positioning coil is installed in the fixed position of the arc-shaped reference object of the magnetic registration module. The subject enters the magnetic shielding barrel and the magnetic detection equipment starts to work.

[0016] Step 3: Activate the high-precision current source to provide a weak high-frequency current signal to the positioning coil of the magnetic registration module, enabling it to operate;

[0017] Step 4: Begin collecting and saving magnetocardiogram (MCC) data;

[0018] Step 5: After the acquisition of the magnetic field signal, the subject wears the CT registration module registration reference, and the wearing position is consistent with the position of the magnetic field registration arc reference, which is on both sides of the body; the subject wears only the CT registration module registration reference for CT scan to obtain CT image structure and save CT data;

[0019] Step 6: Segment and 3D reconstruction of the saved CT data. Register the position of the circular positioning coil in the magnetocardiogram with the position of the registration reference object in the CT image. Then, perform fine registration between the arc-shaped reference object of the magnetocardiogram registration module and the CT reconstruction structure. Use the nearest neighbor point cloud method to improve the registration accuracy.

[0020] Furthermore, in step six, rigid registration is performed using the least squares method based on the reference point. The specific method is as follows:

[0021] Given a point set P = {p1, p2, ..., p...} n}, Q={q1,q2,...,q n Find the rotation matrix R and the translation vector t such that the error function The point set P, which approaches the minimum value, is composed of the position coordinates of the circular positioning coil at the reference object of the magnetic field registration module detected by SERF, and the point set Q is composed of the fixed base coordinates of the circular coil in the registration reference object of the CT registration module obtained by CT scan.

[0022] Furthermore, in step six, the rotation and translation matrices of the two point clouds and the corresponding registration errors are solved using the nearest neighbor point cloud method and Euclidean transformation, including:

[0023] After rigid registration, the data is used for point cloud registration. The error is evaluated using the root mean square error, which is expressed by the following formula:

[0024]

[0025]

[0026] Where, p closest For point p in the registered SERF data point set P i Find the nearest neighboring point in the rigidly registered data point set Q, and solve for the rotation matrix R and translation vector t such that E is minimized.

[0027] The advantages of this invention over existing technologies are:

[0028] (1) This invention combines SERF-based magnetocardiography equipment with CT equipment and builds a registration module for the two equipment to perform image fusion, laying the foundation for three-dimensional magnetocardiography and promoting the development of next-generation magnetocardiography technology.

[0029] (2) This invention designs an integrated and detachable registration module, namely, the magnetic cardiomyography registration module is detachable from the magnetic cardiomyography device, making it applicable to each subject. It can accurately locate the registration markers and obtain the relative position information between the markers and the sensor module. Subsequently, using its own structural components and marker coordinates, it can be registered with the CT three-dimensional reconstruction structure. This system is not affected by human respiration, improving registration accuracy, making the magnetic cardiomyography device more complete, and promoting subsequent three-dimensional imaging applications.

[0030] (3) In this invention, the subject first undergoes a magnetic heart detection, and then wears a CT registration module that is consistent with the magnetic heart device registration module to perform a CT scan. This method can shorten the registration time and improve the registration accuracy, thus promoting the development of magnetic heart three-dimensional imaging technology. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the structure of a magnetic resonance imaging registration system based on CT and SERF according to the present invention;

[0032] Figure 2 This is a schematic diagram of the registration structure of a magnetic cardiomyography registration system based on CT and SERF according to the present invention;

[0033] Figure 3 This is a flowchart of a magnetic resonance imaging registration method based on CT and SERF according to the present invention.

[0034] The labels in the diagram are as follows: 1-Magnetic shielding barrel for the heart magnetometer; 2-Magnetic array panel for the heart magnetometer; 3-SERF magnetometer; 4-Fixed arm for the magnetic array for the heart magnetometer; 5-Arc-shaped reference object for the heart magnetometer registration module; 6-Cylindrical base and annular support column; 7-High-precision current source; Detailed Implementation

[0035] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Example 1

[0037] This invention discloses a CT and SERF-based magnetic resonance imaging (MRI) registration system, comprising a MRI registration module and a CT registration module. The MRI registration module includes a multi-turn circular positioning coil and an adjustable patch limiting device, wherein the positioning coil can be fixed to a protrusion of the limiting device. Figure 1As shown, this invention includes a cardiac magnetic shielding barrel 1 and two arc-shaped reference objects 5 for cardiac magnetic registration modules, distributed on both sides of the body to account for the influence of respiration. Each arc-shaped reference object 5 has three cylindrical bases and annular support columns 6 for fixing and positioning coils. The cylindrical bases and annular support columns 6 are perpendicular to the arc-shaped reference object 5. SERF magnetometers 3 are arranged and inserted into the cardiac magnetometer array panel 2, and the relative position of the SERF magnetometers 3 and the human body is adjusted using the cardiac magnetometer array fixing arm 4. The CT registration module registration reference object is consistent with the arc-shaped reference object 5 of the cardiac registration module, having cylindrical bases and annular support columns 6, both made of resin material. The CT registration module reference object is visible during CT scanning imaging, enabling rapid positioning of reference points. The consistency between the CT registration module reference object and the arc-shaped reference object of the cardiac registration module reduces registration errors and shortens experimental preparation time. By combining CT with the SERF (Self-Regulating Electrocardiogram) device, rapid registration of magnetocardiograms (MCGs) with CT 3D images is achieved for subsequent 3D imaging, promoting the clinical application of MCG. The integrated design allows the MCG registration module to be detached from the MCG device, making it suitable for every subject. This invention can accurately locate the arc-shaped reference object of the MCG registration module, obtaining the relative position information between the reference object and the sensor module. Subsequently, using its own structural components and marked coordinates, it can be registered with the CT 3D reconstruction structure. This system is unaffected by human respiration, improving registration accuracy and making the MCG device more complete, thus advancing subsequent 3D imaging applications.

[0038] The magnetic registration module includes a high-precision current source 7, which generates a weak current signal of stable intensity to power the positioning coil. When the SERF magnetometer 3 detects the magnetic signal, it extracts the positioning coil signal from the signal for positioning.

[0039] The circular positioning coil, as shown Figure 2 As shown, the positioning coil, fixed on a circular base, has four turns, an inner diameter of 5 mm, and a wire diameter of 0.5 mm. A high-precision current source 7 generates a stable alternating current, which is passed through the circular positioning coil to produce a specific magnetic field. The SERF magnetometer 3 can identify and calculate the positional relationship between the coil and the magnetometer.

[0040] The arc-shaped reference object 5 of the magnetic cardiomyography registration module is made of resin, with an arc length of 120mm, a width of 50mm, and a thickness of 5mm. Of the cylindrical base and annular support column 6, the circular base has a diameter of 10mm and a height of 5mm, and the annular support column has a diameter of 5mm and a height of 7mm. The CT module registration reference object is the same size as the arc-shaped reference object 5 of the magnetic cardiomyography registration module. The reference object's density differs from that of human tissue, allowing it to be extracted from the image during CT scanning for three-dimensional reconstruction of the human structure and rapid rigid registration based on the reference object.

[0041] The SERF (Sequential Electrocardiogram-Resistant Response) device comprises a magnetic shielding container 1, a magnetic field array panel 2, a SERF magnetometer 3, a magnetic field array fixing arm 4, and a magnetic registration module. The magnetic shielding container 1 provides an extremely weak magnetic field environment for the SERF magnetometer 3. The SERF magnetometer 3 is inserted and fixed into the magnetic field array panel 2. The magnetic field array fixing arm 4 can move back and forth and up and down to position the center of the magnetic field array panel 2 at the subject's xiphoid process. Adjusting the arc-shaped reference object 5 of the magnetic registration module to fit against both sides of the body, unaffected by respiration, results in more accurate positioning.

[0042] The CT and SERF-based magnetic-cardiac imaging registration system includes a CT device. Before CT imaging, a CT registration module is fixed on both sides of the subject's body as a registration reference object. During CT imaging, the reference object information is acquired. The CT registration module's registration reference object has the same shape as the arc-shaped reference object 5 of the magnetic-cardiac registration module, further improving registration accuracy and reducing registration error.

[0043] The cylindrical base and annular support column 6 are perpendicular to the surface of the arc-shaped reference object 5 of the magnetic registration module, facilitating positioning and marking after CT 3D imaging. Marking is only required at the center of the positioning coil support column, making it quick and easy.

[0044] The positioning coil of the arc-shaped reference object 5 of the magnetic heart registration module and the reference point of the registration reference object of the CT registration module are rigidly registered using the least squares method, and then fine registration is performed based on the nearest point cloud.

[0045] Example 2

[0046] The present invention provides a method for registering magnetic resonance imaging (MRIE) based on CT and SERF, which uses the aforementioned CT and SERF-based MRIE system for MRIE registration. The specific implementation includes the following steps:

[0047] Step 1: The subject lies on the cardiac magnetosurface. The SERF magnetometer 3 is fixed to the cardiac magnetosurface array panel 2. The cardiac magnetosurface array fixing arm 4 is moved so that the middle position of the cardiac magnetosurface array panel 2 is aligned with the xiphoid process of the human body.

[0048] Step 2: Adjust the arc-shaped reference objects 5 of the magnetic registration module on both sides of the magnetic registration bed board to fit the sides of the human body. The positioning coil is assembled on the cylindrical base and annular support column 6 of the arc-shaped reference object 5 of the magnetic registration module. The subject enters the magnetic shielding barrel 1, and the magnetic signal detection module starts to work.

[0049] Step 3: Start the high-precision current source 7 to provide a stable, weak, high-frequency current signal to the positioning coil of the magnetic registration module, so that it can work.

[0050] Step 4: Start the SERF magnetometer 3 to begin collecting magnetocardiogram data and save the relevant information.

[0051] Step 5: After the magnetic resonance imaging (MRI) signal acquisition is completed, the subject wears the CT registration module registration reference, positioned on both sides of the body, consistent with the position of the arc-shaped reference 5 on the MRI registration module. The subject undergoes a CT scan while wearing only the CT registration module registration reference to obtain CT image structures and save the CT data.

[0052] Step 6: Segment and 3D reconstruction of the saved CT data. Register the location of the circular positioning coil in the magnetocardiogram with the location of the reference object in the CT image. Then, finely register the arc-shaped reference object of the reference object registered by the CT registration module with the CT reconstruction structure. Use the nearest neighbor point cloud method to improve the registration accuracy.

[0053] As a preferred method, the reference points are rigidly registered using the least squares method. The specific method is as follows:

[0054] Given a point set P = {p1, p2, ..., p...} n}, Q={q1,q2,...,q n Find the rotation matrix R and the translation vector t such that the error function The point set P, which tends to the minimum value, is composed of the position coordinates of the circular positioning coils at 5 locations of the arc-shaped reference object of the magnetic field registration module detected by SERF, and the point set Q is composed of the fixed base coordinates of the circular coils in the registration reference object of the CT registration module obtained by CT scan detection.

[0055] Using the nearest neighbor point cloud method, the rotation and translation matrices of the two point clouds and the corresponding registration errors are solved through Euclidean transformation. Point cloud registration is then performed on the data after rigid registration, and the root mean square error (RMSE) is used to evaluate the error. Specifically, this can be expressed by the following formula:

[0056]

[0057]

[0058] Where, p closest For point p in the registered SERF data point set Pi Find the nearest neighboring point in the rigidly registered data point set Q, and solve for a new rotation matrix R and translation vector t such that E is minimized.

[0059] In magnetic resonance imaging (MRI) registration, the position of the positioning coil in the MRI registration module and its coordinates relative to the corresponding reference object in CT 3D imaging directly affect registration accuracy. This invention separately acquires data from the MRI and CT registration modules, stores the data in a database, and utilizes rigid registration overlaid with point cloud registration for precise registration, thus improving accuracy. After registration, multimodal data is obtained, which, combined with clinical research and MRI data, is then used for further MRI imaging data processing.

[0060] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0061] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A registration method for a CT and SERF-based magnetic resonance imaging registration system, characterized in that, The CT and SERF-based magnetic cardiac imaging registration system includes a magnetic cardiac registration module and a CT registration module. The magnetic cardiac registration module includes a magnetic cardiac registration module arc-shaped reference and a circular positioning coil located in the magnetic cardiac registration module arc-shaped reference. The CT registration module includes a CT registration module registration reference made of a non-magnetic material with the same shape as the magnetic cardiac registration module arc-shaped reference. The registration method includes the following steps: Step 1: The subject lies on the magnetic cardiopulmonary bypass bed, and the magnetic cardiopulmonary bypass panel is moved so that the center of the panel is aligned with the xiphoid process of the human body; Step 2: Adjust the arc-shaped reference objects of the magnetic registration module on both sides of the magnetic bed to fit the sides of the human body. The positioning coil is installed in the fixed position of the arc-shaped reference object of the magnetic registration module. The subject enters the magnetic shielding barrel and the magnetic detection equipment starts to work. Step 3: Activate the high-precision current source to provide a weak high-frequency current signal to the positioning coil of the magnetic registration module, enabling it to operate; Step 4: Begin collecting and saving magnetocardiogram (MCC) data; Step 5: After the acquisition of the magnetic field signal, the subject wears the CT registration module registration reference, and the wearing position is consistent with the position of the magnetic field registration arc reference, which is on both sides of the body; the subject wears only the CT registration module registration reference for CT scan to obtain CT image structure and save CT data; Step 6: Segment and 3D reconstruction of the saved CT data. Register the position of the circular positioning coil in the magnetocardiogram with the position of the registration reference object in the CT image. Then, perform fine registration between the arc-shaped reference object of the magnetocardiogram registration module and the CT reconstruction structure. Use the nearest neighbor point cloud method to improve the registration accuracy. In step six, rigid registration is performed using the least squares method based on the reference points. The specific method is as follows: Known point set , Find the rotation matrix R and the translation vector t such that the error function The set of points tends to the minimum value. The point set consists of the position coordinates of the circular positioning coil at the reference object of the magnetic field registration module detected by SERF. The coordinates of the fixed base of the circular coil in the registration reference object of the CT registration module are obtained by CT scan detection.

2. The registration method according to claim 1, characterized in that, The magnetic registration module includes a high-precision current source that provides a stable, high-frequency, weak current signal to the circular positioning coil.

3. The registration method according to claim 1, characterized in that, The arc-shaped reference object of the magnetic registration module is made of resin material, and has a cylindrical base and annular support column on top to fix the positioning coil.

4. The registration method according to claim 1, characterized in that, The magnetic registration module consists of two parts, located on either side of the subject's torso, and is unaffected by respiratory movements.

5. The registration method according to claim 1, characterized in that, The device includes a magnetocardiogram (MCC) detection device, which includes a magnetocardiogram (MCC) magnetometer array panel, a magnetocardiogram (MCC) bed, and a magnetocardiogram (MCC) shielding barrel. The MCC magnetometer array panel can be adjusted in height and angle, and the MCC bed is equipped with a magnetocardiogram (MCC) registration module with an arc-shaped reference object.

6. The registration method according to claim 1, characterized in that, The system includes CT detection equipment. Before CT imaging, CT registration modules are fixed on both sides of the subject's body as registration reference objects. During CT imaging, information about the CT registration modules and registration reference objects is collected.

7. The registration method according to claim 1, characterized in that, The arc-shaped reference object of the magnetic cardiography registration module and the registration reference object of the CT registration module are registered using a reference point and nearest neighbor algorithm.

8. The registration method according to claim 1, characterized in that, In step six, the nearest neighbor point cloud method is used to solve for the rotation and translation matrices of the two point clouds and the corresponding registration errors through Euclidean transformation, including: After rigid registration, the data is used for point cloud registration. The error is evaluated using the root mean square error, which is expressed by the following formula: in, For the registered SERF data point set Points in Data point set after rigid registration Find the nearest neighboring points, and solve for the rotation matrix R and translation vector t such that... It is the minimum value.

Images