Multi-fiber laser ablation system and method guided by magnetic resonance

Abstract

The invention provides a magnetic resonance guided multi-fiber laser ablation system and method, the system comprises a host, the host stores an executable instruction, and when the executable instruction is executed by the host, the position of a target area and scanning parameters are determined; performing fast magnetic resonance imaging on the target region periodically, and obtaining a group of continuous magnetic resonance images in each period; for each group of magnetic resonance images, a three-dimensional temperature diagram is established according to phase diagrams in the magnetic resonance images, and information support is provided for the multi-fiber laser ablation process. According to the method, the position and the scanning parameters of the target area are accurately determined, fast magnetic resonance imaging in the operation is carried out according to the scanning parameters, the three-dimensional temperature diagram is established, under the condition that the real-time requirement is met, three-dimensional temperature data are obtained, a user can conveniently determine the temperature state of each position of the target area, and the user experience is improved. And on the basis, simultaneous ablation of one or more optical fibers is performed, so that the operation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly relates to a magnetic resonance-guided multi-fiber ablation system and method. Background Art

[0002] Magnetic resonance-guided laser interstitial thermotherapy (MRgLITT) is a method of using a laser to ablate a target lesion. It uses an optical fiber to introduce light energy into the target tissue. As the irradiation time increases, the temperature of the target tissue rises, and finally reaches the temperature required for thermal ablation, thereby eliminating the lesion. During the operation, a magnetic resonance device is used to detect the temperature state of the target tissue to ablate the target area as much as possible while avoiding ablation of normal tissue. Magnetic resonance-guided laser interstitial thermotherapy has the advantages of minimally invasive, low risk, and fast recovery, etc., and provides new treatment options for various diseases (such as brain tumors, epilepsy, etc., liver, spleen, prostate tumors, etc.).

[0003] Since the light-emitting part at the end of the optical fiber emits laser light to the target tissue in a direction substantially perpendicular to the optical fiber, in order to better observe the ablation state, the current common solution is to set the scanning parameters of the magnetic resonance device so that the magnetic resonance device acquires magnetic resonance images at several mutually parallel cross-sections perpendicular to the optical fiber direction and passing through the target tissue for generating a temperature map. The reason for collecting magnetic resonance images only at several cross-sections passing through the target area is to increase the update frequency of the temperature state and meet the real-time requirement of temperature monitoring.

[0004] The problem faced by the prior art is that sometimes the range of the target area is large (for example, the diameter is greater than 3 cm) or the shape of the target area is irregular, and it is difficult to completely ablate the target area through a single optical fiber (by adjusting the depth and direction of the optical fiber to adjust the depth and direction of the light-emitting part at the end). At this time, it is necessary to plan multiple paths and use multiple optical fibers from different paths to achieve complete ablation of the target area. Since the magnetic resonance device can only collect magnetic resonance images in the direction of one optical fiber each time, it is convenient to view the corresponding temperature state and evaluate the ablation effect of the optical fiber at the corresponding cross-section. If multiple optical fibers are used for ablation simultaneously, there will be a temperature monitoring blind area and a risk of mis-ablation. Therefore, currently, laser ablation can only be performed sequentially under each optical fiber path. After the laser ablation under one optical fiber path is completed, medical staff need to enter the magnetic resonance room or push the patient back to the operating room, implant the optical fiber according to the next planned path, and then perform the laser ablation of this path in the magnetic resonance room. The above process requires multiple interruptions of the operation and multiple entries into the sterile environment, increasing the risk of patient infection, and the laser ablation of the optical fiber under multiple paths is performed sequentially, resulting in low surgical efficiency.

[0005] For the above-mentioned defects, the present invention provides a magnetic resonance-guided multi-fiber laser ablation system and method. Summary of the Invention

[0006] The present invention provides a magnetic resonance-guided multi-fiber laser ablation system and method to solve the defects in the prior art, reduce the infection risk of patients, and shorten the operation time.

[0007] The present invention provides a magnetic resonance-guided multi-fiber laser ablation system, including a host. The host stores executable instructions, and when the executable instructions are executed by the host, the following functions are realized:

[0008] Determine the position of the target area and the scanning parameters;

[0009] Periodically perform rapid magnetic resonance imaging on the target area, and obtain a set of continuous magnetic resonance images in each period; wherein, the set of continuous magnetic resonance images includes the image information of one or more optical fibers;

[0010] For each set of magnetic resonance images, establish a three-dimensional temperature map based on the phase map in the magnetic resonance image to provide information support for the multi-fiber laser ablation process.

[0011] According to the magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the determination of the position of the target area and the scanning parameters includes:

[0012] Obtain the positioning sequence scanning image, establish a positioning three-dimensional structure diagram based on the positioning sequence scanning image, and register the positioning three-dimensional structure diagram with the preoperative three-dimensional image to determine the position of the target area;

[0013] According to the position of the target area, determine the scanning parameters; or, determine the scanning parameters according to the user input instruction.

[0014] According to the magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the determination of the scanning parameters according to the target area in the positioning three-dimensional structure diagram includes:

[0015] Determine the directions of each optical fiber after actual implantation according to the positioning three-dimensional structure diagram, and use the cross-section perpendicular to the first optical fiber direction as the scanning direction; wherein, the first optical fiber direction is the optical fiber direction determined according to the user input instruction, or the optical fiber direction randomly determined from the directions of each optical fiber after actual implantation;

[0016] Determine the number of scanning layers according to the size of the target area and the scanning resolution in the positioning three-dimensional structure diagram.

[0017] According to the magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the three-dimensional image includes at least one of the following parameters:

[0018] The target area, key points / regions to be protected, and the planned implantation paths of one or more optical fibers.

[0019] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, which periodically performs fast magnetic resonance imaging on the target area, includes:

[0020] Determine the target area to be sampled in the current period according to a preset magnetic resonance downsampling rule;

[0021] Perform magnetic resonance scanning on the target area to obtain a magnetic resonance image of the target area;

[0022] Obtain a continuous magnetic resonance image of the current period according to the magnetic resonance image of the target area.

[0023] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, wherein the preset magnetic resonance downsampling rule is inter-slice downsampling and / or intra-slice downsampling.

[0024] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, wherein obtaining a continuous magnetic resonance image of the current period according to the magnetic resonance image of the target area includes:

[0025] Obtain a continuous magnetic resonance image of the current period by interpolation according to the magnetic resonance image of the target area; or,

[0026] Obtain a continuous magnetic resonance image of the current period according to the magnetic resonance image of the target area in combination with the magnetic resonance image of its complementary period.

[0027] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, wherein periodically performing fast magnetic resonance imaging on the target area includes:

[0028] Periodically scan the target area through a magnetic resonance fast imaging sequence;

[0029] Correct the magnetic resonance image of the current period using a reference period to obtain a corrected magnetic resonance image.

[0030] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, wherein the magnetic resonance image of the reference period is a positioning sequence scan image obtained by pre-scanning a reference area that covers the target area.

[0031] A magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, wherein the magnetic resonance fast imaging sequence is a gradient echo sequence, and the positioning sequence is a spin echo sequence.

[0032] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the reference period is any corrected period, and the correction of the magnetic resonance image of the current period using the reference period includes:

[0033] Obtain distortion correction information according to the first position of the target tissue in the magnetic resonance image of the reference period and the second position of the target tissue in the magnetic resonance image of the current period; wherein, the reference period is a positioning sequence scan image obtained by pre-scanning a reference area, and the reference area covers the target area;

[0034] Correct the magnetic resonance image of the current period according to the distortion correction information.

[0035] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the host computer executing the executable instructions further realizes:

[0036] Obtain a control instruction, and control an actuator to perform a corresponding action according to the control instruction to achieve simultaneous ablation of one or more optical fibers; wherein, the control instruction is an instruction input by a user, or a default instruction included in the executable instruction, or an automatically generated instruction.

[0037] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the actuator includes at least one of the following devices:

[0038] An optical fiber robot for adjusting at least one of the following parameters according to a corresponding control instruction: the depth of the optical fiber, the angle, the moving direction of the optical fiber, the moving speed of the optical fiber;

[0039] A laser for adjusting at least one of the following parameters according to a corresponding control instruction: laser power, duty cycle, output duration;

[0040] A peristaltic pump for adjusting the circulation rate of the coolant according to a corresponding control instruction.

[0041] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the host computer executing the executable instructions further realizes:

[0042] Fuse and display the three-dimensional temperature map with the preoperative three-dimensional image;

[0043] According to a cross-section display instruction input by a user, display a corresponding cross-section effect in the fused image of the three-dimensional temperature map and the preoperative three-dimensional image; and / or,

[0044] According to a cross-section adjustment instruction input by a user, adjust and display a corresponding cross-section effect in the fused image of the three-dimensional temperature map and the preoperative three-dimensional image.

[0045] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, when the host executes the executable instructions, the following is further implemented:

[0046] Judge the ablation state of the tissue according to each phase three-dimensional temperature map and the corresponding acquisition time.

[0047] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, when the host executes the executable instructions, the following is further implemented:

[0048] Calculate the currently ablated area according to the three-dimensional temperature map, and display the currently ablated area in the preoperative three-dimensional image.

[0049] According to a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention, the system further includes: a magnetic resonance device, a laser ablation component, and a cooling component.

[0050] The present invention also provides a data processing method for a magnetic resonance-guided multi-fiber laser ablation system, which is characterized by including:

[0051] Determine the position of the target area and the scanning parameters;

[0052] Periodically perform fast magnetic resonance imaging on the target area, and obtain a set of continuous magnetic resonance images in each period; wherein, the set of continuous magnetic resonance images contains the image information of one or more optical fibers;

[0053] For each set of magnetic resonance images, establish a three-dimensional temperature map according to the phase map in the magnetic resonance images, and fuse and display the three-dimensional temperature map with the preoperative three-dimensional image to provide information support for the multi-fiber laser ablation process.

[0054] The magnetic resonance-guided multi-fiber laser ablation system and method provided by the present invention have the following beneficial effects:

[0055] 1. Based on fast magnetic resonance imaging, a three-dimensional temperature map is reconstructed, and more comprehensive temperature data is generated while meeting the real-time requirement, which is convenient for observing the temperature state at each position in combination with the tissue structure.

[0056] 2. It can simultaneously monitor the laser ablation states of multiple optical fibers, provide more comprehensive and accurate information support for simultaneous multi-fiber laser ablation, facilitate doctors to perform multi-fiber laser ablation simultaneously, improve the accuracy and safety of laser ablation, and comprehensively improve the surgical efficiency.

[0057] 3. Through the positioning sequence scanning, a three-dimensional structure diagram for positioning the target area of the patient is reconstructed. On this basis, the scanning parameters are reasonably set to accurately cover the target area, laying a foundation for intraoperative fast magnetic resonance imaging, improving the quality of the three-dimensional temperature map and the acquisition efficiency.

[0058] 4. It can not only perform magnetic resonance image acquisition in a direction perpendicular to a certain optical fiber, but also perform image acquisition from any scanning direction, improving the flexibility of image acquisition.

[0059] 5. The host controls the actuator to perform corresponding actions to cooperate with or assist the doctor to achieve "multi-fiber laser ablation simultaneously", improving the accuracy and intelligence of the surgery.

[0060] 6. Process the magnetic resonance image data according to the cross-section display / cross-section adjustment instruction input by the user, and display the corresponding cross-section effect on the display device, facilitating the user to observe the temperature state at various positions and directions, evaluate the ablation state, and improving the accuracy and convenience of temperature monitoring.

[0061] 7. Calculate the ablated area based on the three-dimensional temperature map, and fuse and display the ablated area with the preoperative three-dimensional image, facilitating the user to intuitively understand the position and size of the currently ablated area, and more accurately guiding the laser ablation process.

[0062] 8. Rapidly obtain magnetic resonance images through inter-layer downsampling / intra-layer downsampling, and obtain continuous magnetic resonance images through the "repair" method, shortening the imaging cycle and meeting the real-time requirement of temperature monitoring during the laser ablation process.

[0063] 9. Scan the target area through a magnetic resonance rapid imaging sequence, and correct the obtained magnetic resonance images using the reference cycle to obtain continuous magnetic resonance images, shortening the imaging cycle and meeting the real-time requirement of temperature monitoring during the laser ablation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0065] Figure 1 is one of the structural schematic diagrams of a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention;

[0066] Figure 2 is another structural schematic diagram of a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention;

[0067] Figure 3 is the flowchart of the data processing method of a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0069] The following will describe Figures 1 - 3 a magnetic resonance-guided multi-fiber laser ablation system and method of the present invention.

[0070] Figure 1 is one of the schematic structural diagrams of a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention. As Figure 1 shown, the system 100 includes a host 110; the host 110 can adopt a general computer architecture, in which executable instructions are stored. When the executable instructions are executed by the host 110, the following are realized:

[0071] S11. Determine the position of the target area and the scanning parameters.

[0072] Specifically, in order to perform laser interstitial thermotherapy, it is necessary to determine the position of the target site (the target site is the human body part where the ablation operation is to be performed, such as the liver, spleen, prostate, head, etc.) in the operating room space, and determine the position of the target area to be ablated in the target site in the operating room space, so as to accurately monitor the temperature of the target area (and its surrounding tissues) through a magnetic resonance device. The operating room space can be described by coordinate systems such as the magnetic resonance device coordinate system and the hospital bed coordinate system.

[0073] For example, the method of determining the position of the target area and the scanning parameters is to perform a global positioning sequence scan on the patient through a magnetic resonance device. Since the scanning range of the positioning sequence scan is large enough to cover the target site, the position of the target site and the position of the target area to be ablated included in the target site can be determined according to the positioning sequence scan image. Then, according to the position of the target area, the scanning parameters (such as scanning direction, number of scanning layers, scanning position, scanning resolution, etc.) are determined. Or the scanning parameters are determined in combination with preoperative images and preoperative planning data based on the preoperative images. After the host 110 determines the scanning parameters, the scanning parameters are sent to the magnetic resonance device so that the magnetic resonance device can perform fast magnetic resonance imaging on the target area during the subsequent intraoperative ablation process and monitor the target area and its surrounding tissues.

[0074] During the intraoperative ablation process, we only focus on the temperature status of the target area (and its surrounding tissues), without the need for a global magnetic resonance scan covering the target site. The intraoperative magnetic resonance scan range only covers the target area (and its surrounding tissues), which is smaller than the scan range of the localization sequence scan. Therefore, the intraoperative magnetic resonance imaging cycle can be shortened.

[0075] S12. Rapid magnetic resonance imaging of the target area is performed periodically, and a set of consecutive magnetic resonance images is obtained in each cycle; among them, the image information of one or more optical fibers is included in a set of consecutive magnetic resonance images.

[0076] Specifically, the magnetic resonance device performs rapid magnetic resonance imaging on the target area periodically according to the scanning parameters. Among them, the image data collected in each cycle includes a set of consecutive magnetic resonance images of the target area and its surrounding tissues. Here, "consecutive" means that the magnetic resonance scanning images of each layer are continuously adjacent, and there is no "missing area" between layers. Each set of magnetic resonance images can be used to establish a three-dimensional temperature map and a three-dimensional structure map of the current scanning cycle, facilitating real-time updating of the monitoring data. Here, the rapid magnetic resonance imaging is a magnetic resonance imaging method that can efficiently collect and be used to generate a three-dimensional temperature map. For example, a magnetic resonance imaging method that takes measures such as shortening the repetition time TR, collecting fewer phase-encoding lines, using parallel acquisition technology, using an echo-planar imaging sequence, using a gradient echo imaging sequence, using EPI, PRESTO, b-SSFP imaging sequence, etc.

[0077] Rapid magnetic resonance imaging can obtain magnetic resonance image data with higher resolution and more layers, facilitating the establishment of a three-dimensional temperature map. While meeting the real-time requirements of intraoperative magnetic resonance temperature monitoring, it can also contain more temperature data. For example, in the prior art, 3 layers of magnetic resonance images are collected in each cycle, and the data resolution of each layer (i.e., the slice thickness of each layer) is 3 mm, and the acquisition time of each cycle of data is about 4 s. In an embodiment of the present invention, a rapid magnetic resonance sequence is used for scanning, 20 layers of magnetic resonance images are collected in each cycle, the data resolution of each layer is 1 mm, and the acquisition time of each cycle of data is 5 - 6 s. The data resolution is higher, and it can still meet the real-time requirements. Generally, rapid magnetic resonance imaging refers to a magnetic resonance scan with a data acquisition cycle less than 6 s.

[0078] After the magnetic resonance device has collected the rapid magnetic resonance images, the host 110 can receive and obtain the above data from the magnetic resonance device for data processing. Since the magnetic resonance-guided multi-fiber laser ablation system of the present invention can be used for simultaneous ablation of multiple optical fibers, before intraoperative magnetic resonance temperature monitoring, one or more optical fibers have been implanted in the target area. Correspondingly, the "one or more optical fibers" can be observed in a set of consecutive magnetic resonance images obtained by a cycle of scanning.

[0079] S13. For each set of magnetic resonance images, a three-dimensional temperature map is established based on the phase map in the magnetic resonance images, providing information support for the multi-fiber laser ablation process.

[0080] Specifically, by obtaining magnetic resonance images of continuous thin layers through fast magnetic resonance imaging and establishing a three-dimensional temperature map based on this, the signal-to-noise ratio of the images can be improved, facilitating the observation of temperature changes in three-dimensional space, better guiding the ablation process, and evaluating the ablation status. Each magnetic resonance image contains a phase map and an amplitude map. The amplitude map shows the tissue structure information, and the phase map can be used to generate the temperature map of this layer. The phase maps in a set of magnetic resonance images can be used to establish the three-dimensional temperature map corresponding to this cycle.

[0081] Since the three-dimensional temperature map is collected and reconstructed through fast magnetic resonance imaging, the temperature monitoring blind area is avoided, and the temperature monitoring effect is improved. Doctors can observe the temperature status at each position from any angle and cross-section, understand the ablation status at each position, making the temperature monitoring no longer limited to the cross-section perpendicular to the fiber direction. Ablation of one or more fibers can be performed simultaneously without interrupting the surgical process, improving the surgical efficiency.

[0082] In this embodiment, the position of the target area and the scanning parameters are accurately determined, facilitating the accurate acquisition of magnetic resonance images for the target area during the intraoperative ablation process. By obtaining fast magnetic resonance imaging to establish a three-dimensional temperature map, three-dimensional temperature data are obtained while meeting the real-time requirements, facilitating the user to determine the temperature status at each position in the target area, and based on this, simultaneous ablation of one or more fibers is performed, improving the surgical efficiency.

[0083] Based on any of the above embodiments, in one embodiment, S11 includes:

[0084] Obtain the positioning sequence scan images, establish a positioning three-dimensional structure diagram based on the positioning sequence scan images, and register the positioning three-dimensional structure diagram with the preoperative three-dimensional image to determine the position of the target area;

[0085] Determine the scanning parameters according to the position of the target area; or, determine the scanning parameters according to the user input instruction.

[0086] Specifically, before intraoperative acquisition, a positioning sequence scan is also required to determine the location of the target area. The positioning sequence scan is a large-scale magnetic resonance scan of the patient, used to locate the position and direction of the lesion structure (target area). The positioning sequence scan can be T1WI (T1-weighted imaging), T2WI (T2-weighted imaging), DWI (diffusion-weighted imaging), SE (spin echo sequence imaging) scan, etc. Preferably, the imaging method of the positioning sequence scan image is consistent with that of the preoperative three-dimensional image to improve the registration accuracy between the positioning three-dimensional structure diagram and the preoperative three-dimensional image. The positioning sequence scan image contains a series of amplitude maps, and based on this series of amplitude maps, a three-dimensional structure diagram (i.e., the positioning three-dimensional structure diagram) for positioning the target site can be reconstructed. In this three-dimensional structure diagram, the target site (the human body part to be subjected to ablation operation, such as the liver, spleen, prostate, head, etc.), the target area in the target site, the actual implantation path of the optical fiber, etc. can be observed.

[0087] Since the medical image of the target area has been acquired preoperatively to establish the preoperative three-dimensional image, the doctor has outlined the target area on the preoperative three-dimensional image, or the target area has been determined by automatically semantically segmenting the preoperative three-dimensional image using a semantic segmentation model. On this basis, by registering the positioning sequence scan image with the preoperative three-dimensional image, the position of the target area in the positioning three-dimensional structure diagram can be determined; of course, the target area position can also be directly outlined or automatically semantically segmented based on the positioning sequence scan image.

[0088] According to the position of the target area in the positioning three-dimensional structure diagram, scanning parameters can be determined. The scanning parameters include, for example, the scanning direction, the number of scanning layers, the scanning position, the scanning resolution, etc. Among them, the scanning direction is the direction of the tomogram, the number of scanning layers is the number of magnetic resonance images collected in each scanning cycle to cover the target area (such as 10 layers, 15 layers, 18 layers, etc.), the scanning position is the collection position of the multi-layer magnetic resonance images, and the scanning resolution is the slice thickness of each layer of magnetic resonance images (such as 1mm, 2mm, 3mm, etc.). The scanning parameters can be set automatically. For example, the number of scanning layers is automatically set according to the size of the target area and the scanning resolution; some parameters can also be set randomly (for example, the scanning direction is set randomly); the scanning parameters can also be determined according to the user's input instructions. For example, the user selects the scanning direction that is convenient for observing a certain specific structure according to the organizational structure to be observed. Further, a certain margin of the number of scanning layers can also be set so that the scanning range of the magnetic resonance includes the target area, which is convenient for monitoring the current ablation boundary and determining whether the normal tissue around the target area is accidentally ablated.

[0089] In this embodiment, a three-dimensional structural map for positioning is reconstructed through positioning sequence scanning and matched with the preoperative three-dimensional image, so as to accurately determine the position of the target part and target area, and then reasonably set the scanning parameters on this basis, laying the foundation for rapid magnetic resonance imaging during surgery and improving the quality of the three-dimensional temperature map.

[0090] Based on any of the above embodiments, in one embodiment, determining the scanning parameters according to the position of the target area includes:

[0091] Determine the directions of the optical fibers after actual implantation according to the positioning three-dimensional structure diagram, and use the section perpendicular to the first optical fiber direction as the scanning direction; wherein the first optical fiber direction is the optical fiber direction determined according to the user input instruction, or is a randomly determined optical fiber direction from the directions of the optical fibers after actual implantation;

[0092] The number of scanning layers is determined based on the size of the target area in the three-dimensional structural map and the scanning resolution.

[0093] Specifically, the positioning three-dimensional structure diagram represents the scanned structural information within the scanning range of the magnetic resonance device positioning sequence, and the positioning three-dimensional structure diagram includes the target area and each optical fiber actually implanted in the target area. When automatically determining the scanning parameters, one of the multiple optical fibers can be randomly selected, and the section perpendicular to the direction of the optical fiber can be used as the scanning direction, or the user can select and determine an optical fiber, and the section perpendicular to the direction of the optical fiber can be used as the scanning direction.

[0094] When automatically determining the number of scanning layers, it is necessary to combine the size of the target area to be scanned and the scanning resolution (i.e., layer thickness) to determine it. It is understandable that the larger the target area, the larger the number of scanning layers required; the smaller the scanning layer thickness, the larger the number of scanning layers required. Furthermore, a certain scanning layer margin can be set to ensure that the scanning range of the magnetic resonance completely covers the target area, or covers a certain range around the target area.

[0095] In this embodiment, the scanning parameters of the magnetic resonance are automatically and accurately determined, the intelligence level is improved, and the quality of rapid magnetic resonance imaging is also improved.

[0096] In addition, it should be noted that the above-mentioned section perpendicular to the direction of the first optical fiber is used as the scanning direction, which is convenient for observing the ablation state of the first optical fiber, for example, pre-planning the optical fiber with the largest ablation range. Since the present invention uses rapid magnetic resonance imaging and reconstructs a three-dimensional temperature map, the temperature state can be observed from any direction and angle, so it can also be set arbitrarily, that is, the magnetic resonance scanning direction is not limited to the direction perpendicular to a certain optical fiber.

[0097] Based on any of the above embodiments, in one embodiment, automatically determining scanning parameters includes:

[0098] According to the shape of the target area, the direction of the thinnest sandwich layer is used as the scanning direction; wherein, the thinnest sandwich layer refers to two parallel planes that sandwich the target area range, such that the distance between the planes is minimized.

[0099] Specifically, the position of the target area represents the actual space occupied by the target area, that is, the position of the target area contains the shape information of the target area. When the target area is sandwiched by two parallel planes, when the distance between the two parallel planes is minimized, the "thinnest sandwich layer" is formed. Using the direction of the thinnest sandwich layer as the scanning direction minimizes the overall scanning range of the intraoperative magnetic resonance scanning and the number of scanning layers. Correspondingly, when the temperature monitoring requirements can be met, the magnetic resonance scanning cycle can be shortened and the update frequency of the three-dimensional temperature map can be increased.

[0100] In this embodiment, according to the shape of the target area, the direction of the thinnest sandwich layer is used as the scanning direction, optimizing the magnetic resonance scanning parameters, improving the magnetic resonance image acquisition rate and the update frequency of the three-dimensional temperature map, while improving the temperature monitoring effect and meeting the real-time requirements of the clinic.

[0101] Based on any of the above embodiments, in one embodiment, the preoperative three-dimensional image includes at least one of the following parameters:

[0102] The target area, key points / regions to be protected, and the planned implantation paths of one or more optical fibers.

[0103] Specifically, the preoperative planning parameters may include the target area manually outlined or segmented after input into the deep learning model. The preoperatively outlined target area can be directly mapped to the positioning three-dimensional structure diagram through registration, facilitating the determination of the target area position and saving surgical time; the preoperative planning parameters may also include key points / regions to be protected, such as brain functional areas (language area, motor area), nerves, blood vessels, etc., which can be displayed in the three-dimensional temperature map through registration to avoid irreversible consequences caused by incorrect ablation of this point / region; the preoperative planning parameters may also include the planned implantation paths of one or more optical fibers, such as the number of optical fibers, the type of optical fiber, the intracranial entry point of the optical fiber, the intracranial entry angle, etc. By setting the planned path, it can provide guidance for the implantation of multiple optical fibers. And after the positioning sequence scanning, combined with the actual optical fiber implantation position, the implantation error of the optical fiber can also be evaluated to determine whether it is necessary to re-implant the optical fiber (adjust the actual implantation path of the optical fiber).

[0104] In this embodiment, by setting each preoperative planning parameter, it provides guidance for the multi-fiber laser ablation process, which helps to improve the accuracy of laser ablation.

[0105] Based on any embodiment, in one embodiment, S12 includes:

[0106] According to the preset magnetic resonance downsampling rule, determine the target area to be sampled in the current cycle;

[0107] Perform a magnetic resonance scan on the target area to obtain a magnetic resonance image of the target area;

[0108] Obtain consecutive magnetic resonance images of the current cycle based on the magnetic resonance image of the target area.

[0109] Specifically, during the intraoperative ablation process, the magnetic resonance device needs to image the target area to be ablated and the tissues nearby it in order to monitor the temperature state. That is, the "consecutive magnetic resonance images obtained per cycle" need to cover the target area. To achieve fast magnetic resonance imaging, when collecting magnetic resonance images in a specific cycle, only a part of the area that "covers the target area" is imaged, and then the "consecutive magnetic resonance images" covering the target area are obtained by means of "supplementary data" repair. The magnetic resonance downsampling rule is a rule preset for determining the target area to be scanned in the current cycle. For example: To completely cover the target area, n layers of magnetic resonance images need to be collected. The preset magnetic resonance downsampling rule is the scanning strategy (i.e., interlayer downsampling) for which layers to scan and which layers not to scan in the current cycle. More specifically, for example, odd-numbered layers are collected in one cycle, and even-numbered layers are collected in the next cycle, and so on; or for example, 3i + 1 layers (i = 0, 1, 2, 3...) are collected in one cycle, 3i + 2 layers (i = 0, 1, 2, 3...) are collected in the next cycle, and 3i + 3 layers (i = 0, 1, 2, 3...) are collected in the next cycle, and so on; or for example, the preset magnetic resonance downsampling rule can also be judged in combination with the number of layers to be sampled. In the case of exceeding the layer threshold, the first strategy is adopted, and in the case of not exceeding the layer threshold, the second strategy is adopted.

[0110] Among them, the layer threshold can be a preset value used to judge whether the number of layers to be scanned in the target area is too large, so as to adopt different interlayer downsampling strategies, such as the layer threshold being 10, 15, 18, etc. The first strategy is the downsampling strategy set for the case where the number of layers to be sampled is too large, and the second strategy is the downsampling strategy set for the case where the number of layers to be sampled is small. Therefore, the layer interval of the first strategy is greater than that of the second strategy. For example, in the case of exceeding 12 layers, it is divided into an odd-numbered layer collection cycle and an even-numbered layer collection cycle. In the case of not exceeding 12 layers, sampling is performed on each layer within one cycle.

[0111] It should be noted that since the scanned target area remains unchanged, there will be the same spatial redundant information between frames. An accelerated acquisition method of interlayer downsampling can be used to accelerate the acquisition of magnetic resonance images. For example, using EPI or GRE sequences, the accelerated acquisition method of interlayer downsampling can improve the image acquisition speed. In this way, only some layers of the target area are scanned within each cycle to accelerate the acquisition of magnetic resonance images, and then the acquired data of the current cycle and the previous cycle can be used to reconstruct a complete image, thereby ensuring the quality of the acquired image. Without changing the image quality and scanning time, high spatio-temporal resolution and large-range temperature imaging can be achieved.

[0112] In an optional implementation manner of this embodiment, the preset magnetic resonance downsampling rule is to sample odd layers and even layers by cycle; according to the preset magnetic resonance downsampling rule, determining the target area to be sampled in the current cycle includes:

[0113] Determining the sampled layers scanned in the previous cycle;

[0114] When the sampled layer in the previous cycle is an odd layer, determining that the target layer to be sampled in the current cycle is an even layer;

[0115] When the sampled layer in the previous cycle is an even layer, determining that the target layer to be sampled in the current cycle is an odd layer.

[0116] That is to say, the layers sampled in two adjacent cycles are different. In practical applications, the sampled layers scanned in the previous cycle can be determined first. If the sampled layer scanned in the previous cycle is an odd layer, then the even layers are sampled in the current cycle. If the sampled layer scanned in the previous cycle is an even layer, then the odd layers are sampled in the current cycle. For example, the continuous magnetic resonance images covering the target area include 1 - 10 layers, one cycle scans layers 1 / 3 / 5 / 7 / 9, and the next cycle samples layers 2 / 4 / 6 / 8 / 10.

[0117] In this embodiment, the area to be sampled is sampled into odd layers and even layers. One cycle samples odd layers, and the other cycle samples even layers. Only some layers of the target area are scanned within each cycle, and the layers sampled in two adjacent areas do not repeat, achieving maximum downsampling, thereby improving the acquisition speed of the magnetic resonance image sequence to meet the clinical requirements for magnetic resonance-based temperature measurement and ablation calculation.

[0118] In an optional implementation manner of this embodiment, the preset magnetic resonance downsampling rule is to adopt the first strategy when the number of layers to be sampled exceeds the layer number threshold, and adopt the second strategy when it does not exceed the layer number threshold; according to the set interlayer downsampling rule, determining the target layer to be sampled in the current cycle includes:

[0119] Obtaining the number of layers to be sampled of the continuous magnetic resonance images;

[0120] When the number of layers to be sampled exceeds the layer threshold, determine the target layer to be sampled in the current period according to the first strategy;

[0121] When the number of layers to be sampled does not exceed the layer threshold, determine the target layer to be sampled in the current period according to the second strategy.

[0122] Specifically, the layer threshold can be a preset value, which is used to judge whether the number of layers to be scanned in the target area is too large, so as to adopt different inter-layer downsampling strategies. For example, the layer threshold can be 10, 15, 18, etc. The first strategy is a downsampling strategy set for the case where the number of layers to be sampled is too large, and the second strategy is a downsampling strategy set for the case where the number of layers to be sampled is small. Therefore, the layer interval of the first strategy is greater than that of the second strategy. That is to say, due to the excessive number of layers to be sampled in the target area, the first strategy ignores more layers and does not sample them in one period to ensure the scanning speed. For example, the first strategy can sample every k - 1 layers in different periods. That is, in the first period, sample the first layer, in the next period, sample the (1 + k)th layer, and in the next period, sample the (1 + 2k)th layer, and so on; the second strategy can sample every m - 1 layers in different periods, where m is less than k.

[0123] In specific implementation, the first strategy or the second strategy can be configured based on actual requirements. The first strategy or the second strategy can also sample odd and even layers respectively at the above-mentioned interval periods.

[0124] Exemplarily, assume that the target area includes 1 - 12 layers to be sampled, and the layer threshold is 10. At this time, the number of layers to be sampled in the target area exceeds the layer threshold. In one period, scan layers 1 / 4 / 7 / 10, in the second period, sample layers 2 / 5 / 8 / 11, and in the third period, collect layers 3 / 6 / 9 / 12. Additionally, assume that the layer threshold is 15. At this time, the number of layers to be sampled in the target area does not exceed the layer threshold. At this time, one can sample layers 1 / 3 / 5 / 7 / 9 / 11 in one period and layers 2 / 4 / 6 / 8 / 10 / 12 in the next period. That is to say, when the number of layers to be sampled in the target area exceeds the layer threshold, sample every 2 layers in each period; when the number of layers to be sampled in the target area does not exceed the layer threshold, sample every 1 layer in each period.

[0125] It should be noted that when the number of layers to be sampled exceeds the layer threshold, it indicates that the number of layers to be sampled in the target area is relatively large. At this time, the first strategy can be used to determine the target layer to be sampled in the current period. When the number of layers to be sampled does not exceed the layer threshold, it indicates that the number of layers to be sampled in the target area is relatively small, and the second strategy is used to determine the target layer to be sampled in the current period. For different situations of the target area, different inter-layer downsampling strategies can be adopted respectively, which can adapt to more application scenarios and improve applicability and flexibility.

[0126] In an alternative implementation of this embodiment, in addition to using the inter-layer downsampling method to improve the scanning speed, the intra-layer downsampling method can also be used to further improve the scanning speed to meet the requirements of temperature monitoring. Fast magnetic resonance imaging scans by columns. According to the preset magnetic resonance downsampling rule, the target area to be sampled in the current period is determined, including:

[0127] Determine the data columns to be sampled in the current period according to the preset intra-layer downsampling rule;

[0128] Perform magnetic resonance scanning on the data columns to be sampled to obtain a magnetic resonance image sequence of the target area.

[0129] It should be noted that the magnetic resonance fast imaging sequence scans by columns, such as GRE. That is, when scanning a certain layer, the magnetic resonance fast imaging sequence does not scan the entire layer at once, but scans a column of data in a layer at a time. For the magnetic resonance fast imaging sequence that scans by columns, in addition to the above-mentioned inter-layer downsampling, intra-layer downsampling can also be used, and only some columns in a layer are collected in one scan to improve the scanning speed. Additionally, it should be noted that if the magnetic resonance fast imaging sequence scans by layer, such as EPI, when scanning a certain layer, it scans the entire layer at once, and intra-layer downsampling cannot be performed, only inter-layer downsampling can be performed.

[0130] Specifically, the set intra-layer downsampling rule is a pre-set data column selection strategy for intra-layer data columns, that is, for a certain layer, which data columns to collect and which data columns not to collect. For example, the set intra-layer downsampling rule can be to collect the 2nth column data in the current period, the (2n + 1)th column data in the next period, the (2n + 2)th column data in the next period, and so on; or, use the method of 3n, 3n + 1, 3n + 2 to increase the data collection speed within the layer by 2 times or 3 times.

[0131] In practical applications, the set inter-layer downsampling rule is to select some column data from the column data of each layer for collection. According to the set intra-layer downsampling rule, it can be determined which data columns to be collected in the current period, and then the magnetic resonance fast imaging sequence is used to scan the data columns to be collected in the target layer to obtain the magnetic resonance image of the current period. In this way, for the target layer, only some data columns need to be collected, greatly improving the data collection speed within the layer.

[0132] It should be noted that for the target area (i.e., the ablation area), due to the continuity of temperature in physical space and time, the temperature information of adjacent positions can be used to estimate the temperature of the unsampled part, thus ensuring the temperature imaging quality of the target area.

[0133] In specific implementation, all samplings can be pre - performed on the test area based on the temperature - measuring optical fiber to obtain a complete image through testing, and a down - sampled image is obtained by down - sampling the test area. Then, the complete image obtained through testing is used as a label, and the down - sampled image is used as the input of the deep - learning model to train the deep - learning model, so as to obtain a deep - learning model capable of reconstructing a complete image. Subsequently, the data sequence collected by down - sampling is input into the trained deep - learning model, and a reconstructed complete image sequence can be obtained.

[0134] In the embodiments of this specification, an accelerated acquisition method of inter - layer down - sampling and intra - layer down - sampling can be adopted. Only partial layers of the target area and partial data columns in these partial layers are scanned in each period to accelerate the acquisition of magnetic resonance images. Then, a complete image is reconstructed based on the collected data. When reconstructing the image, a deep - learning reconstruction algorithm is used. Utilizing the invariance of the spatial information of the target area and the continuity of temperature in the physical space and time, the complete image information is reconstructed, thereby ensuring the quality of the acquired image and realizing high spatio - temporal resolution and large - range temperature imaging without changing the image quality and scanning time.

[0135] It should be noted that in actual application scenarios, the methods of inter - layer down - sampling and intra - layer down - sampling can be used simultaneously to accelerate the acquisition of magnetic resonance images, or the method of inter - layer down - sampling or intra - layer down - sampling can be used alone to accelerate the acquisition of magnetic resonance images. The embodiments of this specification do not limit this. For example, the layers to be sampled in the target area can be determined, and each layer to be sampled is used as the layer to be scanned in the current period. Then, for each layer, based on the intra - layer down - sampling rule, it is determined which data columns are to be collected.

[0136] Based on any one of the embodiments, in one embodiment, obtaining the continuous magnetic resonance images of the current period according to the magnetic resonance image of the target area includes:

[0137] Obtaining the continuous magnetic resonance images of the current period by interpolation according to the magnetic resonance image of the target area; or,

[0138] Obtaining the continuous magnetic resonance images of the current period by combining the magnetic resonance image of the target area with the magnetic resonance image of its complementary period.

[0139] Specifically, for example, the preset magnetic resonance down - sampling rule is to sample odd - numbered layers and even - numbered layers separately. One period samples 1 / 3 / 5 / 7 / 9 layers, and the next period samples 2 / 4 / 6 / 8 / 10 layers, and so on in a cycle. At T nMagnetic resonance data of layers 1 / 3 / 5 / 7 / 9 were collected periodically, and the data of layers 2 / 4 / 6 / 8 / 10 were missing. The magnetic resonance images of the even layers can be directly filled in by interpolation based on the magnetic resonance images of the odd layers. For example, based on the magnetic resonance images of layer 1 and layer 3, the magnetic resonance image of layer 2 can be filled in, and based on the magnetic resonance images of layer 3 and layer 5, the magnetic resonance image of layer 4 can be filled in until continuous magnetic resonance images of 10 layers are obtained. Or, the missing data can also be filled in according to the complementary period. For example, on the basis of the magnetic resonance data of layers 1 / 3 / 5 / 7 / 9 collected in period T n the magnetic resonance data of the even layers collected in period T n-1 are also combined to fill in the missing even-layer data in period T n or the magnetic resonance data of the even layers collected in complementary periods such as T n-3 and T n-5 are also combined to fill in the missing even-layer data in period T n . The accuracy of the data filled in directly through the interpolation step is limited. Generally, the structure / temperature state of the patient's target tissue does not change suddenly. Therefore, the data in the complementary period of the current period can be used as a reference for the "interpolation process" of the current period to further improve the repair accuracy. It can be understood that the above only exemplifies the "repair" process in the case of interlayer downsampling with an interval of one layer. In the case of a larger interval between layers, the missing layers can also be filled in by linear interpolation (or further combined with the complementary period), and the missing data can also be repaired in the same way in the case of intralayer downsampling.

[0140] In this embodiment, continuous magnetic resonance images are obtained by interpolation based on the downsampled magnetic resonance images, which improves the efficiency of collecting magnetic resonance images, meets the requirement of real-time temperature monitoring during the laser ablation process, and further combines complementary periods to fill in the missing magnetic resonance data, improving the accuracy of the obtained continuous magnetic resonance images.

[0141] Based on any one of the embodiments, in one embodiment, S12 includes:

[0142] Scanning the target area periodically by a magnetic resonance fast imaging sequence;

[0143] Correcting the magnetic resonance images of the current period by using a reference period to obtain corrected magnetic resonance images.

[0144] Specifically, the magnetic resonance fast imaging sequence can be echo planar imaging (EPI), and the obtained magnetic resonance images are the EPI sequence. Among them, EPI is one of the fastest magnetic resonance imaging methods so far. It is a special form of gradient echo. It uses a fast reverse gradient to generate a series of gradient echoes within a single relaxation time (TR) and phase-encode them respectively, filling them into the corresponding k-space to achieve cross-sectional imaging.

[0145] The fast magnetic resonance imaging sequence can also be obtained by scanning the target area based on other fast scanning algorithms, such as shortening the repetition time TR, collecting fewer phase-encoding lines, parallel acquisition technology, gradient echo magnetic resonance pulse sequence (GRE), gradient echo translation imaging (PRESTO), balanced steady-state free precession imaging (B-SSFP), etc.

[0146] The fast magnetic resonance imaging sequence can often complete the acquisition of an image in a few milliseconds. However, during the process of rapidly acquiring the image, information may be lost, resulting in distortion of the acquired image, and further causing the temperature map constructed subsequently to be inaccurate, affecting the operation. Taking the echo planar imaging (EPI) as an example of the fast magnetic resonance imaging sequence, EPI can complete the acquisition of an image in a few milliseconds. Therefore, this technology is one of the fast signal acquisition methods, and rapid scanning of 10 - 20 layers of images (1 - 4 s) can be achieved through the fast EPI sequence. However, since the EPI sequence generates signals through continuous alternating changes of gradients and there is no refocusing pulse to correct the inhomogeneity of the main magnetic field and the error of phase information, as the EPI factor increases, the accumulated phase error becomes larger and larger, and the related deformation and artifacts of the image will also become heavier.

[0147] In practical applications, since the position of the patient or the phantom remains unchanged during the process of scanning the target area through the fast magnetic resonance imaging sequence, the tissue spatial information of the same scanning plane remains unchanged, such as blood vessels, tumor boundaries, cerebral sulci and gyri, etc., that is, the spatial distribution of the target area remains basically unchanged. Therefore, the present invention uses the invariant spatial position information to correct the distortion generated during the fast scanning process, thereby realizing the acquisition of a fast 3D temperature sequence without reducing the spatial resolution. Specifically, the reference period without spatial distortion (or that has completed distortion correction) can be used to correct the obtained magnetic resonance image sequence of the current period, and the corrected magnetic resonance image of the current period can be obtained, thereby realizing efficient and accurate three-dimensional temperature detection.

[0148] Based on the previous embodiment, in one embodiment, the reference period is the positioning sequence scan image obtained by pre-scanning the reference area, and the reference area covers the target area.

[0149] Specifically, the localization sequence scan can be T1WI (T1-weighted imaging), T2WI (T2-weighted imaging), DWI (diffusion-weighted imaging), SE (spin echo sequence imaging) scan, etc. The scope of the localization sequence scan is relatively large. For example, when laser ablation is required for an intracranial tumor, a localization sequence scan is performed on the patient's head (reference area). Based on the localization sequence scan, the location of the tumor (target area) can be determined, which facilitates rapid magnetic resonance imaging of a small area of the tumor (target area) through a magnetic resonance device during the laser ablation process, improving the real-time performance of temperature detection. It can be understood that the imaging scope of the above-mentioned localization sequence scan images is large and the imaging accuracy is high. In this embodiment, it is also used as a reference period to correct the magnetic resonance rapid sequence images collected during the operation and repair the distortion therein. Specifically, using the reference period to correct the magnetic resonance image of the current period to obtain a corrected magnetic resonance image includes:

[0150] Obtain the localization sequence scan images obtained by pre-scanning;

[0151] Correct the magnetic resonance image of the current period according to the localization sequence scan images to obtain a corrected magnetic resonance image.

[0152] In a possible implementation, the third position information of the target tissue (such as blood vessels, tumor boundaries, cerebral sulci, etc.) in the magnetic resonance image of the current period can be determined first, and the fourth position information of the target tissue in the localization sequence scan images can be determined. Then, a spatial transformation matrix is calculated based on the third position information and the fourth position information, and the spatial transformation matrix is used as distortion correction information to correct the magnetic resonance image sequence of the current period to obtain the corrected magnetic resonance image of the current period. In another implementation, the localization sequence scan images and the magnetic resonance image sequence of the current period can be directly compared to correct the magnetic resonance image of the current period to obtain the corrected magnetic resonance image of the current period, and at the same time, the distortion correction information is output and stored for subsequent use.

[0153] In this embodiment, directly correcting the magnetic resonance images obtained in one period with the localization sequence scan images can eliminate the distortion generated during the intraoperative magnetic resonance rapid imaging sequence scan, ensure the accuracy of the magnetic resonance images, and thus ensure the accuracy of the subsequent constructed three-dimensional temperature map.

[0154] Based on any embodiment, in one embodiment, the reference period is any corrected period. Using the reference period to correct the magnetic resonance image of the current period includes:

[0155] Obtain distortion correction information based on the first position of the target tissue in the magnetic resonance image of the reference period and the second position of the target tissue in the magnetic resonance image of the reference period; wherein, the reference period is the positioning sequence scan image obtained by pre-scanning the reference area, and the reference area covers the target area.

[0156] Correct the magnetic resonance image of the current period according to the distortion correction information.

[0157] Specifically, in this embodiment, the reference period is the period of performing a positioning sequence scan on the reference area, and the reference period is the corrected period. During the process of correcting the magnetic resonance image of the current period, first obtain the first position of the target tissue (such as blood vessels, tumor boundaries, cerebral sulci, etc.) in the reference period and the second position of the target tissue in the reference period, calculate the spatial transformation matrix according to the first position and the second position, and use this spatial transformation matrix as the distortion correction information to correct the magnetic resonance image sequence of the current period to obtain the corrected magnetic resonance image of the current period.

[0158] In this embodiment, the distortion correction is not directly performed according to the positioning sequence scan image, which reduces the difficulty of calculating the distortion correction information and improves the data processing efficiency.

[0159] Based on any embodiment, in one embodiment, correcting the magnetic resonance image of the current period according to the reference period to obtain the corrected magnetic resonance image of the current period includes:

[0160] Obtain the distortion correction information of the previous period according to the reference period and the previous period of the current period;

[0161] Use the distortion correction information to correct the magnetic resonance image of the current period to obtain the corrected magnetic resonance image of the current period;

[0162] Based on the magnetic resonance image of the current period and the positioning sequence scan image, determine and store the distortion correction information of the current period, wherein the distortion correction information of the current period is used to correct the magnetic resonance image of the next period of the current period.

[0163] It should be noted that the distortion correction information of the previous period can be obtained first according to the reference period and the previous period of the current period, and then the distortion correction information of the previous period can be directly used to correct the magnetic resonance image of the current period to obtain the corrected magnetic resonance image of the current period, and then the distortion correction information of the current period can be determined for use in the next period of correction. That is to say, each period is corrected based on the distortion correction information of the previous period.

[0164] As an example, after the reference cycle, at the second cycle, since the previous cycle of the second cycle is the reference cycle, at this time, based on the positioning sequence image of the reference cycle and the magnetic resonance image of the second cycle, the magnetic resonance image of the second cycle can be corrected to obtain the corrected magnetic resonance image sequence of the second cycle, and at the same time, the distortion correction information of the second cycle is obtained and stored; then, at the third cycle, the distortion correction information of the second cycle can be directly obtained to correct the magnetic resonance image of the third cycle to obtain the corrected magnetic resonance image sequence of the third cycle, and then based on the positioning sequence scan image of the reference cycle and the magnetic resonance image of the third cycle, the distortion correction information of the third cycle can be calculated and stored; then, at the fourth cycle, the distortion correction information of the third cycle can be directly obtained to correct the magnetic resonance image of the fourth cycle, and so on.

[0165] In this embodiment, each cycle is corrected based on the distortion correction information of the previous cycle, which can reduce the accumulation of distortion information in multiple cycles and improve the accuracy of correction.

[0166] In an optional implementation manner of this embodiment, according to the magnetic resonance image and the positioning sequence scan image of the current cycle, the distortion correction information of the current cycle is determined and stored, including:

[0167] Determine the fifth position information of the target tissue in the magnetic resonance image of the current cycle;

[0168] Determine the sixth tissue position information of the target tissue in the positioning sequence scan image;

[0169] According to the fifth position information and the sixth position information, calculate the spatial transformation matrix, and use the spatial transformation matrix as the distortion correction information and store it.

[0170] Specifically, the target tissue is a tissue with an unchanged position in the patient or the phantom during the scanning process, such as blood vessels, tumor boundaries, cerebral sulci, etc.

[0171] It should be noted that the fifth position information of the target tissue in the magnetic resonance image of the current cycle can be determined through image analysis, and the second tissue position information of the target tissue in the positioning sequence scan image can be determined, and then the spatial transformation matrix is calculated according to the first tissue position information and the second tissue position information. This spatial transformation matrix is the distortion correction information between the first tissue position information and the second tissue position information. In this way, the position information of the invariant target tissue is used to determine the distortion correction information of the current cycle and store it. At the next cycle, the distortion correction information of the current cycle can be directly obtained for correction.

[0172] In an alternative embodiment of this example, in addition to the method where the magnetic resonance images obtained in each cycle are directly corrected based on the positioning sequence scan images scanned in the reference cycle, or each cycle is directly corrected based on the distortion correction information of the previous cycle, any cycle can be optionally selected as a constant reference cycle to correct the magnetic resonance images of each cycle. Specifically, correcting the magnetic resonance image of the current cycle according to the reference cycle to obtain the corrected magnetic resonance image of the current cycle includes:

[0173] Select any cycle as the constant reference cycle;

[0174] Obtain the constant distortion correction information according to the reference and constant reference cycles;

[0175] Use the constant distortion correction information to correct the magnetic resonance image of the current cycle to obtain the corrected magnetic resonance image of the current cycle.

[0176] It should be noted that any cycle can be selected as the constant reference cycle, then the constant distortion correction information is obtained according to the reference cycle and the constant reference cycle, and then the obtained constant distortion correction information is used to correct the magnetic resonance image of the current cycle to obtain the corrected magnetic resonance image of the current cycle. That is to say, the magnetic resonance images of each cycle are corrected based on the constant distortion correction information, and the constant distortion correction information can be determined based on any selected cycle, and the correction method is more flexible and can adapt to different scenario requirements.

[0177] Among them, for the specific implementation process of obtaining the constant distortion correction information according to the reference cycle and the correction reference cycle, reference can be made to the above specific implementation process of determining the distortion correction information of the current cycle based on the magnetic resonance image of the current cycle and the positioning sequence scan image, which will not be elaborated here.

[0178] In addition, during fast magnetic resonance imaging, if a fast magnetic resonance imaging sequence is used and the patient also has head movement, at this time, it is preferably to correct the magnetic resonance image of the current cycle according to the reference cycle to obtain a series of corrected magnetic resonance images, so as to avoid the defect that the imaging regions of the current cycle and the (non-reference cycle) reference cycle are inconsistent and the correction error caused by the distortion of both is relatively large. Or, directly register the distorted images collected in the reference cycle and the current cycle to obtain the distortion correction information, and use this distortion correction information to correct the distortion correction information of the reference cycle, and then use it to correct the magnetic resonance image of the current cycle. Specifically, correcting the magnetic resonance image of the current cycle using the reference cycle includes:

[0179] Register the magnetic resonance image of the reference cycle with the magnetic resonance image of the current cycle to obtain the distortion correction information;

[0180] According to the distortion correction information and distortion correction information of the reference period, correct the magnetic resonance image of the current period to obtain a continuous magnetic resonance image of the current period.

[0181] The distortion correction information of the reference period can be obtained with reference to the foregoing embodiments. In this embodiment, when there is imaging distortion in the magnetic resonance rapid imaging sequence and the patient also has head movement, the imaging distortion is successfully corrected, the imaging quality is improved, and it is also convenient to obtain more accurate magnetic resonance equipment scanning parameters.

[0182] Based on any of the foregoing embodiments, in one embodiment, the host running the executable instructions further implements:

[0183] S14. Obtain a control instruction, and control an actuator to perform a corresponding action according to the control instruction to simultaneously ablate one or more optical fibers; wherein, the control instruction is an instruction input by a user, or a default instruction included in the executable instruction, or an automatically generated instruction.

[0184] Specifically, the host 110 may receive an instruction input by a user through an input device (such as a mouse, a keyboard, a touch screen, etc.). After the host 110 obtains the control instruction, it controls the actuator to perform a corresponding action to simultaneously ablate one or more optical fibers. Of course, the control instruction of the host 110 may be an instruction input by a user. For example, the control instruction may also be automatically generated by the host 110 running the executable instruction according to real-time data. For example, according to the current temperature state, the current optical fiber power, and the heating rate, comprehensively adjust the power of the optical fiber and the coolant circulation rate to prevent the tissue from being heated beyond the temperature threshold; the control instruction may also be a pre-stored instruction in the host 110. For example, read the pre-stored "coolant circulation rate" and control the peristaltic pump to run at a corresponding speed so that the cooling in the cooling circulation pipeline reaches the "coolant circulation rate".

[0185] The host 110 controls the actuator to perform a corresponding action according to the control instruction. For example, it controls the fiber robot to adjust the depth and angle of the optical fiber so that the optical fiber performs precise and conformal ablation; for another example, it controls the laser to adjust the laser power and laser release duration output by the laser; for another example, it controls the peristaltic pump of the cooling component to adjust the coolant circulation rate.

[0186] In this embodiment, the host obtains the control instruction and controls the execution device to perform a corresponding action, accurately realizing the simultaneous ablation of one or more optical fibers, avoiding errors caused by human factors, and improving the degree of intelligence and surgical precision.

[0187] In addition, the doctor can also directly manually control the actuator to perform corresponding actions to achieve simultaneous ablation of one or more optical fibers. For example, manually adjusting the depth and angle of the optical fiber; for another example, directly adjusting the output power on the laser; for another example, manually controlling the adjustment button of the peristaltic pump to adjust the circulation rate of the coolant. In this embodiment, some actuators do not have to be communicatively connected to the host 110 (i.e., do not have to be controlled by the host 110); for another example, a foot switch is used to control whether to input laser to the target area through the optical fiber.

[0188] Based on any of the above embodiments, in one embodiment, the actuator includes at least one of the following devices:

[0189] An optical fiber robot for adjusting at least one of the following parameters according to corresponding control instructions: the depth of the optical fiber, the angle, the movement direction of the optical fiber, and the movement speed of the optical fiber;

[0190] A laser for adjusting at least one of the following parameters according to corresponding control instructions: laser power, duty cycle, and output duration;

[0191] A peristaltic pump for adjusting the circulation rate of the coolant according to corresponding control instructions.

[0192] Specifically, the optical fiber robot can adjust the depth of the optical fiber according to the control instructions. More specifically, it can adjust the movement direction (in-depth / extraction), the movement speed, and the movement stroke (displacement) of the optical fiber. The optical fiber robot can also adjust the angle of the optical fiber. Here, the angle refers to the angle of rotation of the optical fiber along its axis. By adjusting the angle of the optical fiber, the light-emitting section of the optical fiber can release laser along the required direction to the target area to achieve precise and conformal ablation of the target area; the laser can adjust the laser power, duty cycle, and laser release duration according to the control instructions to control the ablation range, ablate the target area as much as possible, and avoid ablation of non-target areas; the peristaltic pump can adjust the rotation speed according to the control instructions, thereby adjusting the circulation rate (i.e., flow rate) of the coolant.

[0193] In this embodiment, the host 110 is used to control specific actuators to perform corresponding actions, which is convenient for cooperating with the doctor to complete laser ablation of multiple optical fibers.

[0194] In addition, the number of each type of actuator mentioned above is not limited to one. Assume that m optical fibers are used simultaneously for ablation during the ablation process. For fiber control, a fiber robot can be configured for each optical fiber; for laser control, a laser can be configured, and m optical path switches can be configured. Each optical path switch controls the on / off of the optical path of one optical fiber (indirectly adjusting the laser power and laser release duration). Alternatively, n lasers (n ≤ m) can be configured, and each laser controls the laser power and laser release duration of at least one optical fiber. The redundant optical fibers and the optical fibers in the same group respectively control the on / off of their respective optical paths through optical path switches; for cooling cycle control, a peristaltic pump can be configured to pump coolant to the cooling pipelines of each optical fiber, and a pipeline speed control valve can be configured on each circulation pipeline to control the coolant circulation rate of the branch. Alternatively, p peristaltic pumps (p ≤ m) can be configured, and each peristaltic pump controls the coolant circulation rate of at least one optical fiber. The redundant optical fibers and the optical fibers in the same group share a peristaltic pump, and a pipeline speed control valve is configured on the cooling circulation branch of the optical fibers sharing the peristaltic pump to adjust the coolant circulation rate of the corresponding branch.

[0195] Based on any of the above embodiments, in one embodiment, the host 110 running the executable instructions further implements:

[0196] Fusing and displaying the three-dimensional temperature map with the preoperative three-dimensional image;

[0197] According to the cross-section display instruction input by the user, displaying the corresponding cross-section effect in the fused image of the three-dimensional temperature map and the preoperative three-dimensional image; and / or,

[0198] According to the cross-section adjustment instruction input by the user, adjusting and displaying the corresponding cross-section effect in the fused image of the three-dimensional temperature map and the preoperative three-dimensional image.

[0199] Specifically, fusing the three-dimensional temperature map with the preoperative three-dimensional image and displaying it on a display device, which is convenient for doctors to observe the temperature status of each tissue. The display device can be an independent display device or a device integrated with the host 110. Further, the corresponding color can be rendered in the preoperative three-dimensional image according to the temperature data in the temperature map to visually represent the temperature status at each position.

[0200] The host 110 also receives instructions input by the user through input devices (such as keyboards, mice, touchscreens, etc.), processes the magnetic resonance image data according to the instructions, and then displays it on a display device. The above display device can be an independent display device or a device integrated with the host 110. The host 110 can display the corresponding cross-sectional effect in the fusion map of the three-dimensional temperature map and the preoperative three-dimensional image according to the cross-sectional display instruction input by the user. For example, the fusion map is divided into two parts (the part on the user's viewing side and the part on the non-user's viewing side) according to the cross-section determined by the user input instruction, and the cross-section and the part of the three-dimensional structure diagram on the non-user's viewing side are displayed on the display device. Further, when the user places the operation cursor on the cross-section, the specific dimension value at the landing point can also be presented. The host 110 can also adjust and display the corresponding cross-sectional effect in the fusion map of the three-dimensional temperature map and the preoperative three-dimensional image according to the cross-sectional adjustment instruction input by the user. The cross-sectional adjustment instruction is, for example, the rotation, translation, etc. of the cross-section position.

[0201] In this embodiment, the magnetic resonance image data is processed according to the cross-sectional display / cross-sectional adjustment instruction input by the user, and the corresponding cross-sectional effect is displayed on the display device, which is convenient for the user to observe the temperature state at various positions and directions, evaluate the ablation state, and improve the accuracy and convenience of temperature monitoring.

[0202] Based on any of the above embodiments, in one embodiment, the host running the executable instructions further implements:

[0203] Judge the ablation state of the tissue according to each phase of the three-dimensional temperature map and the corresponding acquisition time.

[0204] Specifically, the temperature of the tissue and the duration of the temperature determine its ablation state (that is, whether denaturation and necrosis occur). For example, when the temperature is lower than 43°C, the tissue will not be ablated no matter how long the duration is. Another example is that when the temperature is between 43°C and 45°C and lasts for more than 10 minutes, the tissue will be ablated. Another example is that when the temperature is between 50°C and 80°C, the tissue will be ablated within a short time. According to each phase of the three-dimensional temperature map and the corresponding acquisition time, the temperature state data of the tissue at each position over time is determined, and then the ablation state of the tissue at each position is evaluated using an evaluation model. Preferably, the Arrhenius model or the CEM43 model is used to evaluate the ablation of the tissue at each position.

[0205] In this embodiment, based on the three-dimensional temperature map established by fast magnetic resonance imaging, combined with the three-dimensional temperature data in the three-dimensional space, the ablation state at each position is accurately and comprehensively evaluated.

[0206] Based on any of the above embodiments, in one embodiment, the host running the executable instructions further implements:

[0207] Calculate the currently ablated area according to the three-dimensional temperature map and display the currently ablated area in the preoperative three-dimensional image.

[0208] Specifically, the target tissue needs to be maintained at different temperatures for different durations to reach the ablation state. For example, it will not ablate no matter how long it is heated below 43°C. When the temperature reaches above 43°C, the higher the temperature, the shorter the ablation time required. Determine the temperature and duration at each position according to the three-dimensional temperature map in each acquisition cycle, thereby judging whether the tissue at each position is ablated, determining the currently ablated area, and combining the registration relationship between the three-dimensional temperature map and the preoperative three-dimensional image, display the ablated area in the preoperative three-dimensional image. Preferably, the ablated area is displayed transparently and blurred in the preoperative three-dimensional image.

[0209] The three-dimensional temperature map in this embodiment contains comprehensive and accurate temperature information, and the calculated currently ablated area is also more accurate. By displaying the currently ablated area in the preoperative three-dimensional image, it is convenient for the user to intuitively understand the position and size of the currently ablated area, and more accurately guide the laser ablation process.

[0210] Based on any of the above embodiments, in one embodiment, the positioning sequence scanning image is an image obtained based on T1WI.

[0211] Specifically, the positioning sequence scanning is to perform a magnetic resonance scan covering the target site for positioning the target site and determining the spatial positions of the target site and its components (such as the target area). The positioning sequence scanning can be based on T1WI imaging. T1 refers to the spin-lattice relaxation time (also known as the longitudinal relaxation time), and T1WI is T1-weighted imaging, which means highlighting the longitudinal relaxation difference of tissues and minimizing the influence of other characteristics such as transverse relaxation on the image. Since within a certain temperature range, the T1 value has a linear relationship with the temperature, acquiring T1WI for the target site can not only position the target site but also measure the initial temperature of the target site accordingly.

[0212] In addition, the positioning sequence scanning image can also use imaging methods such as T2WI imaging, DWI imaging, and SE imaging. T2 is the transverse duration, T2WI means highlighting the transverse relaxation difference of tissues during imaging, DWI imaging means diffusion-weighted imaging, and SE imaging means spin echo sequence imaging.

[0213] In this embodiment, through the positioning sequence scanning based on T1WI, the target area is accurately positioned, which is convenient for determining the scanning parameters and providing guidance for the ablation of the target area.

[0214] Based on any of the above embodiments, in one embodiment, the magnetic resonance fast imaging sequence is based on the gradient echo sequence.

[0215] Specifically, the purpose of using a fast magnetic resonance sequence is to rapidly acquire multi-layer magnetic resonance images to generate a three-dimensional temperature map with a resolution that meets the requirements, while satisfying the real-time demand for temperature monitoring. The fast field echo (FFE) sequence has a higher imaging efficiency than the spin echo sequence. The spin echo sequence has a longer tissue spontaneous relaxation time, with a longer TE (echo time), such as 10 - 15 ms, while the echo time of the gradient echo can be as short as 2 ms.

[0216] In addition, fast magnetic resonance sequence scanning can also be achieved by shortening the repetition time TR, acquiring fewer phase encoding lines, using parallel acquisition techniques, using gradient echo imaging, using EPI (echo planar imaging), PRESTO (gradient echo shifted imaging), and b-SSFP (balanced steady-state free precession imaging) sequences. A brief description of these fast magnetic resonance sequence scanning methods is as follows:

[0217] Parallel acquisition technique: The parallel acquisition technique uses a multi-channel phased array coil to reduce the acquisition density of the limited K-space encoding lines, accelerating the acquisition speed of magnetic resonance images. It can shorten the image acquisition time by 1 - 16 times. On the premise of unchanged acquisition time, it can increase the spatial resolution or the three-dimensional acquisition imaging range, or increase the number of repeated acquisitions to improve image quality and reduce artifacts.

[0218] EPI (echo planar imaging): EPI is a special form of gradient echo. It uses a fast reverse gradient to generate a series of gradient echoes within a single relaxation time and performs phase encoding on them respectively, filling them into the corresponding K-space to achieve cross-sectional imaging.

[0219] PRESTO (gradient echo shifted imaging): An additional negative gradient pulse is added to the layer rotation gradient of the gradient pulse sequence. A positive gradient pulse with an equal area is given before the next excitation pulse. Then the gradient echo will be generated within the next TR time, which is equivalent to shifting the echo generated within the current TR time to the next TR time. This technique corrects the gradient echo shift, and the sequence name is the PRESTO sequence.

[0220] b-SSFP (balanced steady-state free precession imaging): When both the longitudinal magnetization vector and the transverse magnetization vector reach a steady state, this is called steady-state free precession. When reasonable TR, TE, and flip angle are set such that the various echoes (FID, SE, and STE) generated by multiple radiofrequency pulses are just fused into one echo, reaching a balanced state, this gradient echo sequence is called the balanced steady-state free precession sequence (b-SSFP). The signal acquired by B-SSFP is the fused signal of various echoes, and its information density is higher.

[0221] It should be noted that fast magnetic resonance imaging can acquire magnetic resonance images with higher resolution and more layers, which is convenient for subsequent establishment of temperature maps. When meeting the clinical requirements of intraoperative magnetic resonance temperature monitoring, it can also contain more temperature data. For example, currently, 3 layers of magnetic resonance images are acquired per cycle, and the data resolution of each layer (i.e., the slice thickness of each layer) is 3 mm, and the acquisition time of the data per cycle is about 4 - 6 s. If fast magnetic resonance imaging is used, 20 layers of magnetic resonance images can be acquired per cycle, the data resolution of each layer is 1 mm, and the acquisition time of the data per cycle is 3 - 6 s. The data resolution is higher and still can meet the real-time requirements.

[0222] In this embodiment, through the above specific fast magnetic resonance imaging method, magnetic resonance images are acquired quickly and with high quality, which are used to reconstruct a high-resolution three-dimensional temperature map, providing sufficient information support for the multi-fiber laser ablation process guided by magnetic resonance.

[0223] Based on any of the above embodiments, in one embodiment, the system further includes: a magnetic resonance device 120, a laser ablation assembly 130, and a cooling assembly 140.

[0224] Figure 2 This is the second structural schematic diagram of a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention. Specifically, the magnetic resonance device 120 is used to acquire magnetic resonance images of a patient according to the set scanning parameters and the scanning parameters sent by the host 110, and send them to the host 110 for the host 110 to process, providing information guidance for the multi-fiber laser ablation process; the laser ablation assembly 130 is used to perform specific ablation actions according to the control instructions of the host 110 and / or the manual operation of the doctor. The laser ablation assembly 130 may include devices such as a laser, a fiber robot, and a fiber cannula. The above laser is used to generate laser and introduce the laser into the target area through the fiber. The fiber robot is used to adjust the depth, angle, fiber movement direction, and fiber movement speed of the fiber; the cooling assembly 140 is used to cool the fiber implanted in the target area according to the control instructions of the host 110 and / or the manual operation of the doctor to avoid damaging normal tissues. The cooling assembly 140 may include devices such as a cooling liquid container, a cooling liquid recovery container, a cooling circulation pipeline, and a peristaltic pump.

[0225] The present invention also provides a data processing method for a magnetic resonance-guided multi-fiber laser ablation system. Figure 3 This is the flowchart of a data processing method for a magnetic resonance-guided multi-fiber laser ablation system provided by the present invention. As Figure 3 shown, the method includes:

[0226] S11. Determine the position of the target area and the scanning parameters;

[0227] Specifically, in order to perform laser interstitial thermotherapy, it is necessary to determine the position of the target site (the target site is the human body part where the ablation operation is to be performed, such as the liver, spleen, prostate, head, etc.) in the operating room space, and determine the position of the target area to be ablated in the target site in the operating room space, so as to accurately monitor the temperature of the target area (and its surrounding tissues) through a magnetic resonance device. The operating room space can be described by coordinate systems such as the magnetic resonance device coordinate system and the hospital bed coordinate system.

[0228] For example, to determine the target area position and scanning parameters, a global positioning sequence scan is performed on the patient through a magnetic resonance device. Since the scanning range of the positioning sequence scan is large enough to cover the target site, the position of the target site and the position of the target area to be ablated contained in the target site can be determined based on the positioning sequence scan images. Then, according to the position of the target area, the scanning parameters (such as scanning direction, number of scanning layers, scanning position, scanning resolution, etc.) are determined. Or the scanning parameters are determined by combining preoperative images and preoperative planning data based on the preoperative images. After the host 110 determines the scanning parameters, it sends the scanning parameters to the magnetic resonance device so that the magnetic resonance device can perform fast magnetic resonance imaging on the target area during the subsequent intraoperative ablation process and monitor the target area and its surrounding tissues.

[0229] During the intraoperative ablation process, we only focus on the temperature state of the target area (and its surrounding tissues), and there is no need to perform a global magnetic resonance scan covering the target site. The scanning range of the intraoperative magnetic resonance only covering the target area (and its surrounding tissues) is smaller than the scanning range of the positioning sequence scan. Therefore, the intraoperative magnetic resonance scanning cycle can be shortened.

[0230] S12. Perform fast magnetic resonance imaging on the target area periodically, and obtain a set of continuous magnetic resonance images in each cycle; among them, the image information of one or more optical fibers is included in a set of continuous magnetic resonance images;

[0231] Specifically, the magnetic resonance device performs fast magnetic resonance imaging on the target area periodically according to the scanning parameters. Among them, the image data collected in each cycle contains a set of continuous magnetic resonance images of the target area and its surrounding tissues. Here, continuous means that the magnetic resonance scanning images of each layer are continuously adjacent, and there is no "missing area" between layers. Each set of magnetic resonance images can be used to establish a three-dimensional temperature map and a three-dimensional structure map of the current scanning cycle, which is convenient for real-time updating of the monitoring data. Here, the fast magnetic resonance imaging is a magnetic resonance scanning method that can efficiently collect and be used to generate a three-dimensional temperature map. For example, a magnetic resonance scanning method that adopts means such as shortening the repetition time TR, collecting fewer phase encoding lines, using parallel acquisition technology, using an echo planar imaging sequence, using a gradient echo imaging sequence, using EPI, PRESTO, b-SSFP imaging sequence, etc.

[0232] Fast magnetic resonance imaging can acquire magnetic resonance image data with higher resolution and more layers, facilitating the establishment of a three-dimensional temperature map. While meeting the real-time requirements of intraoperative magnetic resonance temperature monitoring, it can also contain more temperature data. For example, in the prior art, 3 layers of magnetic resonance images are acquired per cycle, and the data resolution of each layer (i.e., the slice thickness of each layer) is 3 mm, and the acquisition time of the data per cycle is approximately 4 s. In an embodiment of the present invention, a fast magnetic resonance sequence is used for scanning, 20 layers of magnetic resonance images are acquired per cycle, the data resolution of each layer is 1 mm, and the acquisition time of the data per cycle is 5 - 6 s. The data resolution is higher, and the real-time requirements can still be met. Generally, fast magnetic resonance imaging refers to magnetic resonance scanning with a data acquisition cycle of less than 6 s.

[0233] After the magnetic resonance device acquires the magnetic resonance images, the host 110 can receive and obtain the above data from the magnetic resonance device for data processing. Since the magnetic resonance-guided multi-fiber laser ablation system of the present invention can be used for simultaneous ablation of multiple fibers, before intraoperative magnetic resonance temperature monitoring, one or more fibers have been implanted in the target area. Correspondingly, the "one or more fibers" can be observed in a set of consecutive magnetic resonance images obtained by one-cycle scanning.

[0234] S13. For each set of magnetic resonance images, establish a three-dimensional temperature map based on the phase map in the magnetic resonance images to provide information support for the multi-fiber laser ablation process.

[0235] Specifically, by acquiring continuous thin-slice magnetic resonance images through fast magnetic resonance imaging and establishing a three-dimensional temperature map based on this, the signal-to-noise ratio of the images can be improved, facilitating the observation of temperature changes in three-dimensional space, better guiding the ablation process, and evaluating the ablation status. Each magnetic resonance image contains a phase map and an amplitude map. The amplitude map shows the tissue structure information, and the phase map can be used to generate the temperature map of this layer. The phase maps in a set of magnetic resonance images can be used to establish the three-dimensional temperature map corresponding to this cycle.

[0236] Since the three-dimensional temperature map is acquired and reconstructed through fast magnetic resonance imaging, the temperature monitoring blind area is avoided, and the temperature monitoring effect is improved. The doctor can observe the temperature status at each position from any angle and section, understand the ablation conditions at each position, so that the temperature monitoring is no longer limited to the section perpendicular to the fiber direction, and one or more fibers can be ablated simultaneously without interrupting the surgical process, improving the surgical efficiency.

[0237] In this embodiment, the position of the target area and the scanning parameters are accurately determined, facilitating the accurate acquisition of magnetic resonance images for the target area during the intraoperative ablation process. By obtaining fast magnetic resonance imaging to establish a three-dimensional temperature map, three-dimensional temperature data is acquired while meeting the real-time requirement, facilitating the user to determine the temperature state at each position of the target area, and on this basis, simultaneous ablation of one or more optical fibers is performed, improving the surgical efficiency.

[0238] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A magnetic resonance-guided multi-fiber laser ablation system, comprising a host, characterized in that, The host stores executable instructions, which when executed by the host implement: Determine the location of the target area and scanning parameters; Periodically perform fast magnetic resonance imaging on the target area, and obtain a set of continuous magnetic resonance images in each period; wherein, the set of continuous magnetic resonance images includes the image information of one or more optical fibers; For each set of magnetic resonance images, establish a three-dimensional temperature map based on the phase map in the magnetic resonance image to provide information support for the multi-fiber laser ablation process.

2. The magnetic resonance-guided multi-fiber laser ablation system according to claim 1, wherein The determination of the location of the target area and the scanning parameters includes: Obtain the positioning sequence scan image, establish a positioning three-dimensional structure diagram based on the positioning sequence scan image, and register the positioning three-dimensional structure diagram with the preoperative three-dimensional image to determine the location of the target area; Determine the scanning parameters according to the location of the target area; or, determine the scanning parameters according to the user input instruction.

3. The magnetic resonance-guided multi-fiber laser ablation system according to claim 2, wherein, The determination of the scanning parameters according to the location of the target area includes: Determine the directions of each optical fiber after actual implantation according to the positioning three-dimensional structure diagram, and use the section perpendicular to the first optical fiber direction as the scanning direction; wherein, the first optical fiber direction is the optical fiber direction determined according to the user input instruction, or the optical fiber direction randomly determined from the directions of each optical fiber after actual implantation; Determine the number of scanning layers according to the size of the target area and the scanning resolution in the positioning three-dimensional structure diagram.

4. The magnetic resonance-guided multi-fiber laser ablation system according to claim 2, wherein, The preoperative three-dimensional image includes at least one of the following parameters: Target area, key points / regions to be protected, planned implantation paths of one or more optical fibers.

5. The magnetic resonance three-dimensional temperature detection device according to claim 1, wherein The periodic fast magnetic resonance imaging of the target area includes: Determine the target area to be sampled in the current period according to the preset magnetic resonance downsampling rule; Perform magnetic resonance scanning on the target area to obtain the magnetic resonance image of the target area; Obtain the continuous magnetic resonance images of the current period according to the magnetic resonance image of the target area.

6. The three-dimensional magnetic resonance temperature detection device according to claim 5, wherein The preset magnetic resonance downsampling rule is inter-slice downsampling and / or intra-slice downsampling.

7. The magnetic resonance three-dimensional temperature detection device according to claim 5, wherein The obtaining of the continuous magnetic resonance images of the current period according to the magnetic resonance image of the target area includes: Obtain the continuous magnetic resonance images of the current period by interpolation according to the magnetic resonance image of the target area; or, Obtain the continuous magnetic resonance images of the current period according to the magnetic resonance image of the target area in combination with the magnetic resonance images of its complementary period.

8. The magnetic resonance three-dimensional temperature monitoring device according to claim 2, characterized in that, The periodic fast magnetic resonance imaging of the target area includes: Periodically scan the target area through the magnetic resonance fast imaging sequence; Use the reference period to correct the magnetic resonance image of the current period to obtain the corrected magnetic resonance image.

9. The magnetic resonance three-dimensional temperature monitoring device according to claim 8, characterized in that The reference period is any corrected period, and the use of the reference period to correct the magnetic resonance image of the current period includes: Obtain the distortion correction information according to the first position of the target tissue in the magnetic resonance image of the reference period and the second position of the target tissue in the magnetic resonance image of the reference period; wherein, the reference period is the positioning sequence scan image obtained by pre-scanning the reference area, and the reference area covers the target area; Correct the magnetic resonance image of the current period according to the distortion correction information.

10. The magnetic resonance-guided multi-fiber laser ablation system according to claim 1, wherein, The host's execution of the executable instructions further implements: Obtain a control instruction and control an actuator to perform corresponding actions according to the control instruction to achieve simultaneous ablation of one or more optical fibers; wherein, the control instruction is an instruction input by a user, or a default instruction included in the executable instructions, or an automatically generated instruction; the actuator includes at least one of the following devices: An optical fiber robot for adjusting at least one of the following parameters according to a corresponding control instruction: the depth of the optical fiber, the angle, the movement direction of the optical fiber, and the moving speed of the optical fiber; A laser for adjusting at least one of the following parameters according to a corresponding control instruction: laser power, duty cycle, and output duration; A peristaltic pump for adjusting the circulation rate of the coolant according to a corresponding control instruction.

11. The magnetic resonance-guided multi-fiber laser ablation system according to claim 1, wherein The host's execution of the executable instructions further implements: Fusion-display the three-dimensional temperature map with the preoperative three-dimensional image; According to a cross-section display instruction input by a user, display a corresponding cross-section effect in the fusion map of the three-dimensional temperature map and the preoperative three-dimensional image; and / or, According to a cross-section adjustment instruction input by a user, adjust and display a corresponding cross-section effect in the fusion map of the three-dimensional temperature map and the preoperative three-dimensional image.

12. The magnetic resonance-guided multi-fiber laser ablation system according to claim 1, wherein The host's execution of the executable instructions further implements: Calculate the currently ablated area according to the three-dimensional temperature map and display the currently ablated area in the preoperative three-dimensional image.

13. A data processing method for a magnetic resonance-guided multi-fiber laser ablation system, characterized in that, Include: Determine the position of the target area and scanning parameters; Periodically perform fast magnetic resonance imaging on the target area, and obtain a set of consecutive magnetic resonance images in each period; wherein, the set of consecutive magnetic resonance images contains the image information of one or more optical fibers; For each set of magnetic resonance images, establish a three-dimensional temperature map according to the phase map in the magnetic resonance image to provide information support for the multi-fiber laser ablation process.

Images