A three-dimensional temperature map construction method and a laser ablation system

Abstract

This invention provides a method for constructing a three-dimensional temperature map and a laser ablation system. The method includes: acquiring temperature-related images at several parallel fracture sites; for each pixel to be supplemented in the region to be constructed, searching for available pixels in the vicinity of the pixel in the temperature-related images to determine the supplementary value of the pixel, thus obtaining the constructed region; and generating a three-dimensional temperature map based on the constructed region. This invention reduces data processing by supplementing data in the three-dimensional region to be constructed. By determining the supplementary value based on available pixels in the vicinity of the pixel to be supplemented, compared to supplementing based on surrounding available pixels within a fracture site, this method utilizes more available pixels, has a wider spatial distribution of available pixels, results in more accurate supplementation, and produces a more spatially smooth three-dimensional temperature map.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a method for constructing a three-dimensional temperature map and a laser ablation system. Background Technology

[0002] In medical scenarios involving thermal ablation of target areas (such as confocal ultrasound, laser interstitial hyperthermia, and radiofrequency ablation), monitoring the target area temperature using magnetic resonance imaging (MRI) helps to understand the ablation status and improve surgical safety. However, the procedure involves thermal ablation of a three-dimensional target area, while MRI scans layer by layer. Acquiring MRI images of the entire target area is typically time-consuming and cannot meet the real-time requirements of temperature monitoring. Therefore, MRI scans are usually performed at several slices passing through the target area to improve temperature monitoring efficiency; that is, the "temperature monitoring range" is sacrificed to ensure "temperature monitoring efficiency."

[0003] Several methods for constructing temperature maps have emerged, capable of generating three-dimensional temperature maps of a target region from magnetic resonance images of only a few tomographic sections. (See reference...) Figure 1 For example, existing technology acquires images at three parallel faults (a, b, and c), and then supplements the temperature map in the vertical fault (d) using linear and cubic interpolation based on existing data at the intersections of fault d with faults a, b, and c. One drawback of this temperature construction method is that supplementing other data solely based on existing data in fault d still results in an insufficiently smooth temperature map, especially when switching to other cross-sections for observation. Another drawback is that using linear interpolation to supplement data places higher demands on the scanning interval; the interval between faults a, b, and c cannot be too large. Therefore, it necessitates "setting a limited number of parallel faults more densely" during image acquisition, which limits the temperature monitoring range.

[0004] To address or at least partially address the aforementioned deficiencies, this invention provides a method for constructing a three-dimensional temperature map and a laser ablation system. Summary of the Invention

[0005] This invention provides a method for constructing a three-dimensional temperature map and a laser ablation system to solve the defects of existing technologies, such as insufficient smoothness of the constructed three-dimensional temperature map and high requirements for the acquisition of raw data.

[0006] In a first aspect, the present invention provides a method for constructing a three-dimensional temperature map, comprising:

[0007] Acquire temperature-related images at several parallel faults;

[0008] For each pixel to be supplemented in the region to be constructed, available pixels in the space near the pixel to be supplemented are searched in the temperature-related image to determine the supplement value of the pixel to be supplemented, and the constructed region is obtained.

[0009] A three-dimensional temperature map is generated based on the constructed region.

[0010] Optionally, the region to be constructed is determined in the following way:

[0011] Based on the transformation relationship between the patient's 3D model space and the image space to be constructed, and the target region in the patient's 3D model, the region to be constructed in the image space to be constructed is determined.

[0012] Optionally, the region to be constructed is determined in the following way:

[0013] Pixels with amplitudes greater than a preset threshold in the temperature-related images at each parallel fault are selected as heated pixels.

[0014] Centered on the heated pixel with the largest amplitude, the area covering all heated pixels is determined as the region to be constructed.

[0015] Optionally, for each pixel to be supplemented in the region to be constructed, searching for available pixels in the vicinity of the pixel to be supplemented in the temperature-related image to determine the supplementary value of the pixel to be supplemented, and obtaining the constructed region, includes:

[0016] Based on the number of pixels to be added in the constructed region, create a corresponding number of parallel processing threads on the graphics card;

[0017] Each parallel processing thread determines the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented based on the fast calculation matrix, and determines the supplementary value of the pixel to be supplemented.

[0018] Furthermore, each parallel processing thread determines the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented based on the fast calculation matrix, and determines the supplemented value of the pixel to be supplemented, including:

[0019] For each pixel to be supplemented in the region to be constructed, a preset number of available pixels in its vicinity are determined by the parallel processing thread of the graphics card corresponding to the pixel to be supplemented.

[0020] Based on a preset number of available pixels near the pixel to be supplemented, the parameters of the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented are determined in conjunction with the fast calculation matrix.

[0021] Substitute the parameter and the coordinates of the pixel to be supplemented into the three-dimensional cubic interpolation formula to obtain the supplemented value of the pixel to be supplemented.

[0022] Furthermore, a preset number of available pixels near the pixel to be supplemented are found based on the Manhattan distance search.

[0023] Optionally, the temperature-related image is a temperature difference map, a phase difference map, or a phase map, and generating a three-dimensional temperature map based on the constructed region includes:

[0024] The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base value to the size of the image space to be constructed to obtain a three-dimensional temperature difference map or phase difference map.

[0025] By combining the three-dimensional temperature-related image and the temperature map when it is not heated, a three-dimensional temperature map for the current period is generated.

[0026] Optionally, the temperature-related image is a temperature map, and generating a three-dimensional temperature map based on the constructed region includes:

[0027] The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base temperature and expanded to the size of the image space to be constructed to obtain a three-dimensional temperature map for the current period.

[0028] Optionally, after generating a three-dimensional temperature map based on the constructed region, the method further includes: performing three-dimensional filtering on the generated three-dimensional temperature map to obtain a smoothed three-dimensional temperature map.

[0029] Secondly, the present invention also provides a method for constructing a three-dimensional temperature map, characterized in that it includes:

[0030] Acquire temperature-correlation images at several fault locations; wherein at least one of the temperature-correlation images at several fault locations passes through a heating center;

[0031] For each pixel to be supplemented in the region to be constructed, the supplement value of the pixel to be supplemented is determined by combining the distance between the heated pixels in its vicinity and the heating center, and the constructed region is obtained.

[0032] A three-dimensional temperature map is generated based on the constructed region.

[0033] Optionally, the region to be constructed is determined in the following way:

[0034] Based on the transformation relationship between the patient's 3D model space and the image space to be constructed, and the target region in the patient's 3D model, the region to be constructed in the image space to be constructed is determined.

[0035] Furthermore, the region to be constructed is determined in the following manner:

[0036] Pixels with amplitudes greater than a preset amplitude threshold and distances from the heating center less than a preset distance threshold in the temperature-related images of each fault are selected as heated pixels.

[0037] Using the heating center as the center, determine the area to be constructed that covers each heated pixel.

[0038] Optionally, for each pixel to be supplemented in the region to be constructed, determining the supplementation value of the pixel to be supplemented, based on the distance between the heated pixels in its vicinity and the heating center, to obtain the constructed region, includes:

[0039] Based on the number of pixels to be added in the area to be built, create a corresponding number of parallel processing threads for the graphics card;

[0040] Each parallel processing thread determines the amplitude distance formula corresponding to the pixel to be supplemented based on the distance between the points near the pixel to be supplemented and the heating center, and determines the supplement value of the pixel to be supplemented.

[0041] Further, each parallel processing thread determines the amplitude distance formula corresponding to the pixel to be supplemented based on the distance between points near the pixel to be supplemented and the heating center, and determines the supplement value of the pixel to be supplemented, including:

[0042] For each pixel to be supplemented in the region to be constructed, a preset number of heated pixels are determined by the parallel processing thread corresponding to the pixel to be supplemented, and the distance between each heated pixel and the heating center is calculated.

[0043] Based on the distance between each heated pixel and the heating center and the temperature difference between each heated pixel, the least squares method is used to fit the amplitude distance formula corresponding to the pixel to be supplemented.

[0044] Substitute the distance between the pixel to be supplemented and the heating center into the amplitude distance formula for the pixel to be supplemented to calculate the supplement value of the pixel to be supplemented.

[0045] Furthermore, the preset number of heated pixels near the pixel to be supplemented are found based on Euclidean distance.

[0046] Optionally, the temperature-related image is a temperature difference map, a phase difference map, or a phase map, and generating a three-dimensional temperature map based on the constructed region includes:

[0047] The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base value to the size of the image space to be constructed to obtain a three-dimensional temperature-related image;

[0048] By combining the three-dimensional temperature-related image and the temperature map when it is not heated, a three-dimensional temperature map for the current period is generated.

[0049] Optionally, the temperature-related image is a temperature map, and generating a three-dimensional temperature map based on the constructed region includes:

[0050] The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base temperature and expanded to the size of the image space to be constructed to obtain a three-dimensional temperature map for the current period.

[0051] Optionally, after generating a three-dimensional temperature map based on the constructed region, the method further includes: performing three-dimensional filtering on the generated three-dimensional temperature map to obtain a smoothed three-dimensional temperature map.

[0052] Optionally, after generating a three-dimensional temperature map based on the constructed region, the resolution of the three-dimensional temperature map for the current period is further improved according to the three-dimensional temperature map construction method of any one of the first aspects.

[0053] Thirdly, the present invention also provides an ablation system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the three-dimensional temperature map construction method as described in any of the first aspects above, or implements the steps of the three-dimensional temperature map construction method as described in any of the second aspects above. The ablation system may be a laser ablation system, a confocal ultrasound ablation system, a radiofrequency ablation system, etc.

[0054] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the three-dimensional temperature map construction method as described in any of the first aspects above, or implements the steps of the three-dimensional temperature map construction method as described in any of the second aspects above.

[0055] Fifthly, the present invention also provides a computer program product comprising computer-executable instructions, which, when executed, are used to implement the steps of the three-dimensional temperature map construction method as described in any of the first aspects above, or to implement the steps of the three-dimensional temperature map construction method as described in any of the second aspects above.

[0056] The three-dimensional temperature map construction method and laser ablation system provided by this invention have at least the following beneficial effects:

[0057] 1. High-precision data supplementation is performed only on the local key regions (i.e., the regions to be constructed) of the image space to be constructed, and efficient filling is performed on the regions outside the regions to be constructed where the temperature changes little or remain basically unchanged, which reduces the amount of data processing and improves the construction efficiency of the 3D temperature map. In addition, in some implementations, a fixed-size region to be constructed is determined to simplify the processing flow, while in other implementations, the region to be constructed is dynamically determined to further reduce the amount of data processing.

[0058] 2. The available pixels are searched in three dimensions in the space near the pixel to be supplemented to determine the supplement value of the pixel to be supplemented (i.e., three-dimensional interpolation). Compared with the prior art, which supplements data layer by layer (i.e., two-dimensional interpolation), the data supplemented by the present invention is smoother.

[0059] 3. Parallel processing threads are constructed, and the graphics card is used to supplement the pixels to be supplemented in the area to be constructed in parallel, which improves the efficiency of three-dimensional temperature construction and better meets the needs of real-time temperature monitoring in ablation surgery.

[0060] 4. For scenarios where parallel faults are collected to construct 3D temperature maps, the parameters of the 3D cubic interpolation formula are quickly determined by rapidly calculating the matrix, which greatly reduces the amount of computation. The overall logic is suitable for parallel processing on graphics cards, balancing the efficiency and quality of 3D temperature construction. In addition, the search for available pixels near the pixel to be supplemented based on Manhattan distance improves the search efficiency, and this search method is more suitable for the scenario of "collecting parallel faults".

[0061] 5. It can expand the data volume of an image and upgrade the image to any desired resolution.

[0062] 6. An alternative method for constructing temperature maps is provided, supporting more flexible data acquisition methods, such as acquiring images at several parallel and / or intersecting tomographic locations. Specifically, by searching for available pixels near the pixel to be supplemented and calculating the distance between each point and the heating center, a dedicated amplitude distance formula is determined for the pixel to be supplemented, which is used to determine the supplement value of the pixel to be supplemented. Even in scenarios where available pixels are unevenly distributed, the temperature map is still accurately constructed. Moreover, the above construction method is also suitable for parallel processing on graphics cards, improving the efficiency of constructing three-dimensional temperature maps. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0064] Figure 1This is an example of interpolation used to construct temperature maps in existing technologies;

[0065] Figure 2 This is a flowchart illustrating a method for constructing a three-dimensional temperature map provided by the present invention;

[0066] Figure 3 This is a flowchart illustrating another method for constructing a three-dimensional temperature map provided by the present invention;

[0067] Figure 4 yes Figure 3 Here is an example of an application scenario corresponding to the method. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0069] The following is combined Figures 2-4 This invention describes a method for constructing a three-dimensional temperature map and a laser ablation method system. Figure 2 This is a flowchart illustrating a method for constructing a three-dimensional temperature map provided by the present invention, as shown below. Figure 2 As shown, the method includes:

[0070] S11. Obtain temperature correlation images at several parallel faults;

[0071] Specifically, in thermal ablation scenarios, because the temperature is constantly changing, it is necessary to continuously and cyclically acquire magnetic resonance images to monitor the ablation status of the target area and provide information for medical personnel. This helps prevent abnormal situations such as "excessive ablation temperature causing vaporization" or "the current ablation range exceeding the planned range," ensuring the safety of the ablation process. It is understood that the aforementioned temperature-related images are images of several parallel slices acquired within a single cycle. This method processes the images of each parallel slice within the same cycle (e.g., 4 seconds) to generate a three-dimensional temperature map for that cycle. By cyclically acquiring images and executing this method, near real-time three-dimensional temperature maps can be continuously generated.

[0072] It is understood that "a number" in this invention refers to at least two. Furthermore, the temperature-related image obtained in step S11 can be a temperature map, a temperature difference map, a phase difference map, or a phase map.

[0073] Magnetic resonance temperature imaging has many technical approaches, such as the proton resonance frequency phase subtraction method. This method utilizes the linear relationship between the water proton resonance frequency and temperature within a certain temperature range (-15 to 100℃) to measure temperature. As the temperature increases, the water proton resonance frequency decreases. By calculating the change in the phase of the magnetic resonance image in the heated region, the change in proton resonance frequency is obtained. This change in proton resonance frequency corresponds to the change in temperature, and a temperature map can be generated accordingly. In other words, (magnetic resonance) phase maps, (magnetic resonance) phase difference maps, (magnetic resonance) temperature difference maps, and (magnetic resonance) temperature maps can be interconverted. This method can supplement these (magnetic resonance) temperature-related images and convert them into three-dimensional temperature maps. For example, the acquired magnetic resonance phase map can be used as a temperature-related image. The phase map is supplemented in step S12, and then in step S13, the difference between this phase map and the phase map before heating is calculated to generate a three-dimensional phase difference map. This phase difference map is further converted into a temperature difference map, and then combined with the temperature map before heating to generate a three-dimensional temperature map for the current period.

[0074] For example, spectral imaging utilizes the characteristic that the resonance frequencies of certain molecules do not change with temperature. Using these as reference signals, the absolute temperature information (which can be converted to degrees Celsius) is obtained by subtracting the resonance frequencies of water protons and hydrogen protons within the same voxel. The resulting temperature map serves as the aforementioned temperature-related image for subsequent processing steps. Alternatively, a (magnetic resonance) temperature difference map, (magnetic resonance) phase difference map, or (magnetic resonance) phase map can be generated based on the (magnetic resonance) temperature map as the aforementioned temperature-related image for subsequent processing steps.

[0075] S12. For each pixel to be supplemented in the region to be constructed, search for available pixels in the space near the pixel to be supplemented in the temperature-related image to determine the supplement value of the pixel to be supplemented, and obtain the constructed region.

[0076] Specifically, this invention requires supplementing missing data areas based on temperature-related images at several fault locations, and generating a three-dimensional temperature map on this basis. The aforementioned image space to be constructed refers to the image space containing temperature-related images at each parallel fault location that requires further supplementation of data. The magnetic resonance images at the aforementioned fault locations are acquired, and correspondingly, the pixels in the processed temperature-related images at each fault location are usable pixels that can be used to determine the supplementary values ​​for the pixels to be supplemented. This step S12 only constructs a local area within the image space to be constructed. The temperature in areas outside the "area to be constructed" is relatively stable and can be quickly supplemented by filling in the base value (background value), thereby improving data processing efficiency. Furthermore, step S12 does not supplement the pixel based on the available pixels around it within a certain fault. Instead, for any pixel to be supplemented, the supplement value of the pixel to be supplemented is determined based on the available pixels in its surrounding space. These available pixels can be located in different faults (planes). More available pixel data is used in the process of "determining the supplement value of the pixel to be supplemented". The available pixels are not limited to the two-dimensional plane. Their spatial distribution is wider and more uniform, which makes the three-dimensional smoothing effect of the constructed image better, rather than only having a smoothing effect in the two-dimensional plane of the temperature map.

[0077] S13. Generate a three-dimensional temperature map based on the constructed region.

[0078] Specifically, since the aforementioned steps construct a local area, a complete image needs to be generated further. In one example, the temperature-related image obtained in step S11 is a temperature map. In this case, step S13 directly places the constructed area into the position corresponding to the "area to be constructed," and fills the unheated area outside the "area to be constructed" with the base temperature to obtain the three-dimensional temperature map for the current period. In another example, the temperature-related image obtained in step S11 is a temperature difference map. In this case, step S13 directly places the constructed area into the position corresponding to the "area to be constructed," and fills the unheated area outside the "area to be constructed" with zeros (i.e., the unheated area does not change temperature). Combined with the temperature map when unheated, the three-dimensional temperature map for the current period is generated. In yet another example, the temperature-related image obtained in step S11 is a phase difference map. In this case, step S13 directly places the constructed area into the position corresponding to the "area to be constructed," and fills the unheated area outside the "area to be constructed" with zeros (i.e., the phase difference of the unheated area remains unchanged). Then, the phase difference image is converted into a temperature difference map, and combined with the temperature map when unheated, the three-dimensional temperature map for the current period is generated. In another example, the temperature-related image obtained in step S11 is a phase map. In this case, step S13 directly places the constructed region into the position corresponding to the "region to be constructed", fills the unheated region outside the "region to be constructed" with the basic phase (i.e. the phase difference of the unheated region remains unchanged), and then generates a three-dimensional phase difference map by subtracting it from the phase map before heating. The phase difference image is then converted into a temperature difference map, and combined with the temperature map when it is not heated, a three-dimensional temperature map for the current period is generated.

[0079] This embodiment reduces data processing by supplementing the 3D region to be constructed with additional data. The supplementary value is determined based on available pixels in the vicinity of the pixel to be supplemented. Compared to supplementing within a specific fault line based on surrounding available pixels, this method utilizes more available pixels, has a wider spatial distribution of available pixels, and results in more accurate supplementation. The constructed 3D temperature map also exhibits better spatial smoothness. Furthermore, it should be noted that the method in this embodiment can also upscale the image to any resolution as needed, resulting in a smoother display. In practical implementation, it is only necessary to supplement pixels with an appropriate density according to the target resolution.

[0080] Based on the previous embodiment, in one embodiment, the region to be constructed is determined as follows: based on the transformation relationship between the patient's three-dimensional model space and the image space to be constructed, and the target region in the patient's three-dimensional model, the region to be constructed in the image space to be constructed is determined.

[0081] Specifically, the patient's 3D model space contains relevant tissue models of the patient, such as blood vessels, tumors, and surrounding tissues. When acquiring magnetic resonance images, the image acquisition range needs to pass through the target area (such as the tumor or epilepsy area). The patient model space and the image space to be constructed have a transformation relationship, which can be determined through image registration, surgical navigation, etc. Based on the transformation relationship, the target area in the patient model space can be mapped to the image space to be constructed, thereby determining the area to be constructed. For example, an area covering the target area can be determined as the area to be constructed, or an area covering the target area and extending to a certain extent can be determined as the area to be constructed. Subsequently, step S12 only needs to determine the supplementary values ​​of the pixels to be supplemented within the area to be constructed, while the temperature of the area outside the area to be constructed is relatively stable, and it can be quickly supplemented by filling the base temperature.

[0082] This invention does not process and supplement every pixel in the entire image space to be constructed. Instead, it determines a local area to be constructed based on the patient's 3D model, and processes and supplements the pixels in that local area to improve processing efficiency while ensuring the quality of the construction.

[0083] Based on any embodiment, in one embodiment, the region to be constructed is determined in the following manner:

[0084] Pixels with amplitudes greater than a preset threshold in the temperature-related images at each parallel fault are selected as heated pixels.

[0085] Centered on the heated pixel with the largest amplitude, determine the area to be constructed that covers all heated pixels.

[0086] Specifically, at each parallel fracture point, heated pixels are selected using a preset threshold. The pixel with the largest amplitude among these heated pixels is used as the center to determine a region covering all heated pixels, i.e., the region to be constructed. This region to be constructed is only a local area of ​​the image space to be constructed, which can improve data processing efficiency. It can be understood that, while covering all heated pixels, the smaller the volume of this region (the fewer pixels it contains), the higher the efficiency of constructing this region.

[0087] In one example, a sphere is defined with the point of maximum amplitude as its center and a radius not less than the "maximum Euclidean distance between available pixels and the heating center" as its radius, serving as the region to be constructed. In another example, a cube is defined with the point of maximum amplitude as its center and a cube is determined by searching for the "maximum Manhattan distance between available pixels and the heating center," also serving as the region to be constructed. Furthermore, a certain margin can be added to this region to ensure effective monitoring of the heated area.

[0088] It is understandable that in this embodiment, the area to be constructed covering the available pixels of that cycle is redefined for each cycle, reducing the amount of data processing. In some special cases, such as the initial heating stage of ablation, the area to be constructed may not cover the target area (e.g., tumor, epilepsy area). In contrast, the previous embodiment directly set a fixed area to be constructed, simplifying the processing flow. For example, a fixed area to be constructed is set based on factors such as the lesion range, ablation boundary range, and key points that need to be protected.

[0089] This embodiment does not process and supplement every pixel in the entire image space to be constructed. Instead, it improves processing efficiency by determining the area to be constructed that covers each heated pixel. This ensures the effectiveness of temperature monitoring.

[0090] Furthermore, prior to step S12, the method further includes setting the amplitude of pixels in the temperature-related image that have amplitude values ​​less than the base value to the base value.

[0091] For example, in one example, points with a temperature greater than 38°C are identified as heated pixels, while points with a temperature less than 37°C are set to the base temperature of the background area, 37°C. In another example, points with a temperature difference greater than 1°C are identified as heated pixels, while points with a temperature difference less than 0°C are set to the base temperature (difference) of the background area, 0°C. By setting pixels with amplitudes less than the base temperature to the base temperature, interference from abnormally low-temperature points can be eliminated. Furthermore, points set to the base temperature can also be used as usable pixels for subsequent determination of supplementary values ​​for pixels to be added.

[0092] Based on any embodiment, in one embodiment, S12 includes:

[0093] S121. Based on the number of pixels to be added in the constructed region, create a corresponding number of parallel processing threads for the graphics card.

[0094] S122. Each parallel processing thread determines the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented based on the fast calculation matrix, and determines the supplementary value of the pixel to be supplemented.

[0095] Specifically, unlike existing interpolation algorithms that perform point-by-point interpolation after fitting a curve, this invention constructs a dedicated parallel processing thread for each pixel to be supplemented, and determines a dedicated three-dimensional cubic interpolation formula based on a fast calculation matrix to determine the supplementary value (amplitude) of the pixel. This greatly improves the efficiency of temperature map construction and also enhances the three-dimensional smoothness of the constructed temperature map. Preferably, the spacing between the parallel fractures in the aforementioned "magnetic resonance image at parallel fractures" is equal to improve the efficiency and accuracy of supplementing pixels using the "three-dimensional cubic interpolation formula".

[0096] Based on the previous embodiment, in one embodiment, S122 includes:

[0097] For each pixel to be supplemented in the region to be built, a preset number of available pixels in its vicinity are determined by the parallel processing thread of the graphics card corresponding to that pixel.

[0098] Based on a preset number of available pixels near the pixel to be supplemented, the parameters of the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented are determined by combining the fast calculation matrix.

[0099] Substitute the parameter and the coordinates of the pixel to be supplemented into the three-dimensional cubic interpolation formula to obtain the supplemented value of the pixel to be supplemented.

[0100] Specifically, each parallel processing thread of the graphics card searches for nearby available pixels based on the position of the pixel to be supplemented. It then uses the relative positions of these available pixels and the pixel to be supplemented, combined with a fast calculation matrix, to efficiently determine the parameters of the 3D cubic interpolation formula. In other words, it efficiently determines a specific 3D cubic interpolation formula for that pixel to be supplemented. Substituting the coordinates of the pixel to be supplemented into this formula determines its supplementary value. It is understandable that the size of this fast calculation matrix is ​​related to the number of available pixels selected; multiplying these points by the fast calculation matrix quickly yields the parameters of the 3D cubic interpolation formula.

[0101] Understandably, if increasing resolution is required during temperature map construction, additional pixels need to be added to each parallel section. Furthermore, to simplify the process, every point in the region to be constructed can be treated as a pixel to be added, even if that pixel already has a true value in the temperature-related image. The advantage of this simplification is that it eliminates the need for additional steps to identify pixels in the region that already have true values. Moreover, since the added value determined by the 3D cubic interpolation formula is itself the same as the true value of that pixel, this simplification does not introduce additional errors.

[0102] This embodiment efficiently determines the three-dimensional cubic interpolation formula for each pixel to be supplemented by quickly calculating the matrix. Furthermore, the designed processing flow of "searching for available pixels that meet the requirements by distance in the set of available pixels" and "querying parameters in the quick calculation table" is particularly suitable for parallel computing, which can improve processing efficiency and better meet the real-time requirements of temperature monitoring.

[0103] Based on the previous embodiment, in one embodiment, a preset number of available pixels near the pixel to be supplemented are found based on Manhattan distance.

[0104] Specifically, the calculation process of Manhattan distance is simpler than that of Euclidean distance. When searching for available pixels in the vicinity of the pixel to be supplemented, Manhattan distance can be calculated quickly directly based on the index, thus improving search efficiency.

[0105] Furthermore, in conjunction with the aforementioned embodiments, during the process of determining the region to be constructed, the region to be constructed can also be determined by taking the heated pixel with the largest amplitude as the center and covering all heated pixels within a certain Manhattan distance range. Specifically, the Manhattan distance between each heated pixel and the center can be calculated separately, and the largest Manhattan distance can be used as a threshold. The distance between the boundary of the region to be constructed and the center needs to be greater than or equal to the largest Manhattan distance.

[0106] Based on any embodiment, in one embodiment, the temperature-related image is a temperature difference map, and S13 includes: filling the constructed region into the image space to be constructed to obtain a three-dimensional temperature difference map; and generating a three-dimensional temperature map for the current period based on the three-dimensional temperature difference image combined with the temperature map when it is not heated.

[0107] It is understandable that the aforementioned image space to be constructed refers to the image space containing temperature-related images (temperature difference maps) at each parallel fault location that requires further data supplementation. Step S12 supplements the temperature of a region covering the heated area. The temperature change outside this region (background region) is small, or in other words, the temperature remains basically unchanged. After filling the constructed region into the image space to be constructed, it is also necessary to fill the missing data areas in these background regions with zeros. The resulting three-dimensional temperature difference map is the temperature difference map of the current period relative to the unheated state. By further superimposing this temperature difference map on the unheated temperature map, the three-dimensional temperature map of the current period can be generated.

[0108] In another embodiment, the temperature-related image is a temperature difference map, and S13 includes: filling the area around the constructed region with zeros to expand it to the size of the image space to be constructed, thereby obtaining a three-dimensional temperature difference map; and generating a three-dimensional temperature map for the current period based on the three-dimensional temperature difference map combined with the temperature map when it is not heated.

[0109] Specifically, the difference from the previous embodiment is that the previous embodiment retained data at each parallel fault outside the construction region, which may contain some noise points and small temperature fluctuations at different locations. This embodiment directly fills these with zeros (base values), and the final effect is equivalent to filling with the base temperature before heating. It can be understood that the above processing logic for the temperature difference map also applies to the phase difference map. The only difference is that after generating the three-dimensional phase difference map, it needs to be further converted into a temperature difference map, and then combined with the temperature map before heating to generate the three-dimensional temperature map for the current period.

[0110] Based on any embodiment, in one embodiment, the temperature-related image is a temperature map, and S13 includes: filling the constructed region into the image space to be constructed to obtain a three-dimensional temperature map of the current period.

[0111] It is understood that the aforementioned image space to be constructed refers to the image space containing temperature-related images at each parallel fault that requires further supplementary data. Step S12 supplements the temperature of a region covering the heated area. The temperature change outside this region is small, or in other words, the temperature remains basically unchanged. If the resolution of the constructed image needs to be improved, after filling the constructed region into the image space to be constructed, it is also necessary to supplement the temperature data in the missing data areas of these background regions. Specifically, the temperature values ​​at the corresponding locations can be supplemented based on the temperature map before heating.

[0112] In another embodiment, the temperature-related image is a temperature map, and S13 includes: expanding the surrounding base temperature of the constructed region to the size of the image space to be constructed, thereby obtaining the constructed three-dimensional temperature map.

[0113] Specifically, the difference from the previous embodiment is that the previous embodiment retained data from parallel faults outside the heated area, which may contain some noise points and small temperature fluctuations at different locations. This embodiment directly fills these data with the base temperature (base value), which can be the average temperature of the background area before heating. It is understood that the above processing logic for the temperature map also applies to the phase map. The difference lies only in that after generating the three-dimensional phase map, it is necessary to further combine it with the phase map before heating to generate a three-dimensional phase difference map, then convert it into a three-dimensional temperature difference map, and finally combine it with the temperature map before heating to generate the three-dimensional temperature map for the current period.

[0114] Based on any embodiment, in one embodiment, after S13, the method further includes: performing three-dimensional filtering on the generated three-dimensional temperature map to obtain a smoothed three-dimensional temperature map.

[0115] Specifically, three-dimensional filtering methods include mean filtering, median filtering, Gaussian filtering, and bilateral filtering. Mean filtering replaces each pixel value with the average of its neighboring pixel values, thus achieving smoothing. Median filtering replaces each pixel value with the median of its neighboring pixel values; this filtering does not rely on the average and can remove noise without blurring image edges. Gaussian filtering replaces each pixel value with a weighted average of its neighboring pixel values, with pixels farther from the center having lower weights; the weight distribution follows a Gaussian distribution, and Gaussian filtering can smooth images while preserving more image details. Bilateral filtering replaces each pixel value with a weighted average of its neighboring pixel values; the weights are determined based on the distance between neighboring pixels and the center pixel, as well as the grayscale difference; bilateral filtering can achieve edge smoothing and noise reduction.

[0116] The present invention also provides a laser ablation system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the steps of the three-dimensional temperature map construction method described above.

[0117] Specifically, the aforementioned three-dimensional temperature map construction method runs as an application within the laser ablation system, generating a three-dimensional temperature map to provide information support for the laser ablation process. The laser ablation system can also perform other functions, such as displaying a three-dimensional model of the patient, estimating the current ablation area based on the constructed temperature map and overlaying it onto the patient's three-dimensional model, or overlaying it onto the boundary of the lesion to be ablated.

[0118] The three-dimensional temperature map construction method provided above is applicable to "application scenarios where magnetic resonance images are acquired at several parallel faults". The following describes another three-dimensional temperature map construction method provided by the present invention, which is applicable to "application scenarios where magnetic resonance images are acquired at any fault". The two methods can be referred to each other.

[0119] Figure 3 This is a flowchart illustrating another method for constructing a three-dimensional temperature map provided by the present invention, as shown below. Figure 3 As shown, the method includes:

[0120] S21. Obtain temperature correlation images at several fault locations; wherein, at least one of the temperature correlation images at several fault locations passes through a heating center;

[0121] Specifically, the aforementioned temperature-related images are images of several fault locations acquired within a single period. This method generates a three-dimensional temperature map for that period based on the images of each fault location within the same period. By repeatedly acquiring images and executing this method, three-dimensional temperature maps can be continuously generated. The positional relationships between the faults are unrestricted; for example, the faults can be parallel to each other, or they can be at certain angles and not intersect, or they can intersect each other, or three faults can intersect orthogonally, or... Figure 4 The scene shown is one where "some faults intersect each other, and some faults are parallel to each other".

[0122] Regardless of the positional relationship of the various tomographic slices, at least one slice's temperature-related image passes through a heating center. The location of this heating center affects the accuracy of the subsequent temperature map construction. The heating center can be determined in several ways. For example, if a thermal ablation plan is planned during the preoperative planning stage, determining the location of the heating center, the surgical navigation system can accurately center heating around this location during the procedure, thus directly acquiring the location of the heating center. Another example is setting the acquisition position of a tomographic image to pass through the fiber optic axis (i.e., the fiber optic cable falls within the plane of the tomographic slice). Since the laser output from the distal end of the fiber optic cable heats the target area, its heating center also falls within the tomographic slice. By analyzing the temperature difference map at this slice, the point with the largest temperature difference is determined as the heating center. Yet another example is performing a high-precision, large-area magnetic resonance imaging (MRI) sequence scan on the patient after fiber implantation to determine the location of the laser output from the distal end of the fiber optic cable (the heating center). Subsequent image acquisition ensures that at least one tomographic slice passes through this heating center. The aforementioned temperature-related images can be temperature maps, temperature difference maps, phase difference maps, or phase maps.

[0123] S22. For each pixel to be supplemented in the region to be constructed, the supplement value of the pixel to be supplemented is determined by combining the distance between the heated pixels in its vicinity and the heating center, and the constructed region is obtained.

[0124] Specifically, this invention requires supplementing missing data areas based on temperature-related images acquired at several fault locations, and generating a three-dimensional temperature map on this basis. The aforementioned image space to be constructed refers to the image space containing temperature-related images from each parallel fault location that requires further supplementation of data. The magnetic resonance images at the aforementioned fault locations are acquired, and pixels at each fault location that meet specific conditions are identified as heated pixels, which can be used to determine the supplementary values ​​for the pixels to be supplemented. This step S22 only constructs a local area within the image space to be constructed. The temperature of areas outside the "area to be constructed" is relatively stable, and can be quickly supplemented by filling with base values ​​(background values), thereby improving data processing efficiency. In addition, this step S22 does not supplement based on available pixels around a pixel within a certain fault location, but rather, for any pixel to be supplemented, the supplementary value of the pixel to be supplemented is determined based on the heated pixels in its surrounding space. These available pixels can be located at different fault locations (planes), and their spatial distribution range is wider and more uniform, resulting in a better three-dimensional smoothing effect in the constructed image, rather than just a smoothing effect in the constructed two-dimensional plane.

[0125] S23. Generate a three-dimensional temperature map based on the constructed region.

[0126] Specifically, since the aforementioned steps construct phase maps, phase difference maps, temperature difference maps, or temperature maps for local regions, it is necessary to further combine them with the temperature map before heating to generate a complete three-dimensional temperature map for the current period. It is understandable that if a temperature difference map is constructed for the local region, it can be directly combined with the temperature map before heating to generate a complete three-dimensional temperature map for the current period. If a phase difference map or phase map is constructed for the local region, it needs to be converted into a temperature difference map first, and then combined with the temperature map before heating to generate a complete three-dimensional temperature map for the current period.

[0127] This embodiment reduces the amount of data processing by determining the three-dimensional region to be constructed. It determines the supplementary value based on the heated pixels in the space near the pixel to be supplemented. Compared with supplementing based on the surrounding pixels in a certain fault, the heated pixels used in the process of "determining the supplementary value of the pixel to be supplemented" are not limited to the two-dimensional plane. Their spatial distribution range is wider, the supplementation result is more accurate, and the spatial smoothness of the constructed three-dimensional temperature map is better.

[0128] Based on the previous embodiment, in one embodiment, the region to be constructed is determined in the following manner:

[0129] Based on the transformation relationship between the patient's 3D model space and the image space to be constructed, and the target region in the patient's 3D model, the region to be constructed in the image space to be constructed is determined.

[0130] Specifically, the patient's 3D model space contains models of the patient's relevant tissues, such as blood vessels, tumors, and surrounding tissues. When acquiring magnetic resonance images, the image acquisition range needs to pass through the target region (such as the tumor or epilepsy area). The patient model space and the image space to be constructed have a transformation relationship, which can be determined through image registration, surgical navigation, etc. Based on the transformation relationship, the target region in the patient model space can be mapped to the image space to be constructed, thereby determining the region to be constructed. For example, a region covering the target region can be determined as the region to be constructed, or a region covering the target region and extending to a certain extent can be determined as the region to be constructed. Subsequently, step S22 only needs to determine the supplementary values ​​of the pixels to be supplemented within the region to be constructed, while the amplitude of the region outside the region to be constructed is relatively stable and can be quickly supplemented by filling with the base value.

[0131] This invention does not process and supplement every pixel in the entire image space to be constructed. Instead, it determines a local area to be constructed based on the patient's 3D model, and processes and supplements the pixels in that local area to improve processing efficiency while ensuring the quality of the construction.

[0132] Based on the previous embodiment, in one embodiment, the region to be constructed is determined in the following manner:

[0133] Pixels with amplitudes greater than a preset threshold and distances from the heating center less than a preset distance threshold in the temperature-related images at each fault are selected as heated pixels.

[0134] Using the heating center as the center, determine the area to be constructed that covers each heated pixel.

[0135] Specifically, at each fault location, heated pixels are selected using preset temperature difference and distance thresholds. A region covering all heated pixels is defined, centered on the heating center, as the region to be constructed. Understandably, the smaller the volume of this region (the fewer pixels it contains), the more efficient the construction of that region, while ensuring coverage of all heated pixels. In one example, a sphere is defined with the heating center as its center and a radius not less than the maximum Euclidean distance between the heated pixels and the heating center, serving as the region to be constructed. In another example, a cube is defined with the heating center as its center and a side length twice the maximum Euclidean distance between the heated pixels and the heating center, serving as the region to be constructed. Furthermore, a certain margin can be added to this region to ensure effective monitoring of the heated areas.

[0136] Understandably, in this embodiment, the area to be constructed covering the available pixels of that period is redefined for each cycle, reducing the amount of data processing. In contrast, the previous embodiment directly set a fixed area to be constructed, simplifying the processing flow, for example, by setting a fixed area to be constructed based on factors such as the lesion range, ablation boundary range, and key points that need to be protected.

[0137] This embodiment does not process and supplement every pixel in the entire image space to be constructed. Instead, it improves processing efficiency by determining the area to be constructed that covers each heated pixel. This ensures the effectiveness of temperature monitoring.

[0138] Based on the previous embodiment, in one embodiment, S22 includes:

[0139] S221. Based on the number of pixels to be supplemented in the area to be built, create a corresponding number of parallel processing threads for the graphics card;

[0140] S222. Each parallel processing thread determines the amplitude distance formula corresponding to the pixel to be supplemented based on the distance between the points near the pixel to be supplemented and the heating center, and determines the supplement value of the pixel to be supplemented.

[0141] Specifically, unlike existing interpolation algorithms that perform interpolation point by point after fitting the curve, this invention creates a dedicated parallel processing thread for each pixel on the graphics card, and fits the amplitude distance formula for each pixel to be supplemented separately to determine the supplementary value (amplitude) of the pixel to be supplemented. This greatly improves the efficiency and accuracy of temperature map construction, and also improves the three-dimensional smoothing effect of the constructed temperature map.

[0142] Based on the previous embodiment, in one embodiment, S222 includes:

[0143] For each pixel to be added in the region to be constructed, a preset number of heated pixels are determined by the parallel processing thread corresponding to the pixel to be added, and the distance between each heated pixel and the heating center is calculated.

[0144] Based on the distance between each heated pixel and the heating center and the temperature difference between each heated pixel, the least squares method is used to fit the amplitude distance formula corresponding to the pixel to be supplemented.

[0145] Substitute the distance between the pixel to be supplemented and the heating center into the amplitude distance formula for the pixel to be supplemented to calculate the supplement value of the pixel to be supplemented.

[0146] Specifically, each parallel processing thread of the graphics card searches for nearby heated pixels based on the location of the corresponding pixel to be supplemented. For example, for each pixel to be supplemented, it searches for the 64 nearest heated pixels (heated pixels are pixels whose amplitude is greater than a preset amplitude threshold and whose distance from the heating center is less than a preset distance threshold). Then, based on these points, it uses the least squares method to fit a special amplitude distance formula for the pixel to be supplemented.

[0147] This embodiment uses parallel processing threads on the graphics card to search for heated pixels near each pixel to be supplemented and determines a specific amplitude distance formula for that pixel. This more accurately determines the supplementary value of the pixel to be supplemented, resulting in a better three-dimensional smoothing effect in the subsequently generated temperature map. The above process design is particularly suitable for parallel processing on the graphics card, which can improve processing efficiency and better meet the real-time requirements of temperature monitoring.

[0148] Furthermore, to simplify the process, every point in the region to be constructed can be treated as a pixel to be supplemented, even if that pixel already has a real value in the temperature difference map. The advantage of this simplification is that there is no need to add a process to identify the few pixels in the region to be constructed that already have real values; after processing, the supplementary value can be overwritten based on the real temperature difference of that pixel.

[0149] Based on the previous embodiment, in one embodiment, a preset number of heated pixels near the pixel to be supplemented are found based on Euclidean distance.

[0150] Specifically, the "distance" in the amplitude distance formula is based on Euclidean distance. When searching for heated pixels near the pixel to be supplemented, searching based on Euclidean distance can obtain more effective pixels, making the amplitude distance formula fitted for the pixel to be supplemented more consistent with the temperature environment around the pixel to be supplemented.

[0151] Based on any embodiment, in one embodiment, the temperature-related image is a temperature difference map, and S13 includes: filling the constructed region into the image space to be constructed to obtain a three-dimensional temperature difference map; and generating a three-dimensional temperature map for the current period based on the three-dimensional temperature difference image combined with the temperature map when it is not heated.

[0152] It is understandable that the aforementioned image space to be constructed refers to the image space containing temperature-related images at each fault location that requires further supplementary data. Step S22 supplements the temperature of a region covering the heated area. The temperature change outside this region (the background region) is small, or in other words, the temperature remains basically unchanged. After filling the constructed region into the original image space, it is also necessary to fill the missing data areas in these background regions with zeros. The resulting three-dimensional temperature difference map is the temperature difference map of the current period relative to the unheated state. By further superimposing this temperature difference map on the temperature map of the unheated state, the three-dimensional temperature map of the current period can be generated.

[0153] In another embodiment, the temperature-related image is a temperature difference map, and S23 includes: filling the area around the constructed region with zeros to expand it to the size of the image space to be constructed, thereby obtaining a three-dimensional temperature difference map; and generating a three-dimensional temperature map for the current period based on the three-dimensional temperature difference map combined with the temperature map when it is not heated.

[0154] Specifically, the difference from the previous embodiment is that the previous embodiment retained some data belonging to each fault outside the heated area, which may contain some noise points and small temperature fluctuations at different locations. This embodiment directly fills these with zeros (base values), and the final effect is equivalent to filling in the base temperature before heating. It can be understood that the above processing logic for the temperature difference map also applies to the phase difference map. The only difference is that after generating the three-dimensional phase difference map, it needs to be further converted into a temperature difference map, and then combined with the temperature map before heating to generate the three-dimensional temperature map for the current period.

[0155] Based on any embodiment, in one embodiment, the temperature-related image is a temperature map, and S23 includes: filling the constructed region into the image space to be constructed to obtain a three-dimensional temperature map of the current period.

[0156] It is understood that the aforementioned image space to be constructed refers to an image space containing temperature-related images at several fault locations that requires further supplementation of data. Step S22 supplements the temperature of a region covering the heated area. The temperature change outside this region is small, or in other words, the temperature remains basically unchanged. If the resolution of the constructed image needs to be improved, after filling the constructed region into the image space to be constructed, it is also necessary to supplement the missing data in these background regions. Specifically, the temperature values ​​at the corresponding locations can be supplemented based on the temperature map before heating.

[0157] In another embodiment, the temperature-related image is a temperature map, and S23 includes: expanding the surrounding base temperature of the constructed region to the size of the image space to be constructed, thereby obtaining the constructed three-dimensional temperature map.

[0158] Specifically, the difference from the previous embodiment is that the previous embodiment retained data from parallel faults outside the heated area, which may contain some noise points and small temperature fluctuations at different locations. This embodiment directly fills these data with the base temperature (base value), which can be the average temperature of the background area before heating. It is understood that the above processing logic for the temperature map also applies to the phase map. The difference lies only in that after generating the three-dimensional phase map, it is necessary to further combine it with the phase map before heating to generate a three-dimensional phase difference map, then convert it into a three-dimensional temperature difference map, and finally combine it with the temperature map before heating to generate the three-dimensional temperature map for the current period.

[0159] Based on the previous embodiment, in one embodiment, after S23, the resolution of the three-dimensional temperature map of the current period is further improved according to the first set of three-dimensional temperature map construction methods described above, so as to meet the needs of medical staff for the clarity of the monitoring screen.

[0160] Based on any embodiment, in one embodiment, after S23, the constructed three-dimensional temperature map is further verified, and outliers are corrected. For example, pixels with temperature values ​​lower than the baseline temperature are selected and set to the baseline temperature to avoid abnormally low-temperature pixels misleading the ablation assessment process.

[0161] Based on any embodiment, in one embodiment, after S23, the method further includes: performing three-dimensional filtering on the generated three-dimensional temperature map to obtain a smoothed three-dimensional temperature map.

[0162] Specifically, three-dimensional filtering methods include mean filtering, median filtering, Gaussian filtering, and bilateral filtering. Mean filtering replaces each pixel value with the average of its neighboring pixel values, thus achieving smoothing. Median filtering replaces each pixel value with the median of its neighboring pixel values; this filtering does not rely on the average and can remove noise without blurring image edges. Gaussian filtering replaces each pixel value with a weighted average of its neighboring pixel values, with pixels farther from the center having lower weights; the weight distribution follows a Gaussian distribution, and Gaussian filtering can smooth images while preserving more image details. Bilateral filtering replaces each pixel value with a weighted average of its neighboring pixel values; the weights are determined based on the distance between neighboring pixels and the center pixel, as well as the grayscale difference; bilateral filtering can achieve edge smoothing and noise reduction.

[0163] The present invention also provides a laser ablation system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the steps of the three-dimensional temperature map construction method described above.

[0164] Specifically, the aforementioned three-dimensional temperature map construction method runs as an application within the laser ablation system, generating a three-dimensional temperature map to provide information support for the laser ablation process. The laser ablation system can also perform other functions, such as displaying a three-dimensional model of the patient, overlaying the generated three-dimensional temperature map onto the patient's three-dimensional model to facilitate doctors' understanding of the temperature status of various tissue areas, estimating the current cumulative ablation area based on the constructed three-dimensional temperature map and overlaying it onto the patient's three-dimensional model, or overlaying it onto the boundary of the lesion to be ablated, to facilitate doctors' understanding of the ablation status of various tissue areas.

[0165] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0166] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a three-dimensional temperature map, characterized in that, include: Acquire temperature-related images at several parallel faults; For each pixel to be supplemented in the region to be constructed, available pixels in the space near the pixel to be supplemented are searched in the temperature-related image to determine the supplement value of the pixel to be supplemented, and the constructed region is obtained. A three-dimensional temperature map is generated based on the constructed region.

2. The method for constructing a three-dimensional temperature map according to claim 1, characterized in that, The region to be constructed is determined in the following way: Based on the transformation relationship between the patient's 3D model space and the image space to be constructed, and the target region in the patient's 3D model, the region to be constructed in the image space to be constructed is determined.

3. The method for constructing a three-dimensional temperature map according to claim 1, characterized in that, The region to be constructed is determined in the following way: Pixels with amplitudes greater than a preset threshold in the temperature-related images at each parallel fault are selected as heated pixels. Centered on the heated pixel with the largest amplitude, the area covering all heated pixels is determined as the region to be constructed.

4. The method for constructing a three-dimensional temperature map according to claim 1, characterized in that, For each pixel to be supplemented in the region to be constructed, the available pixels in the vicinity of the pixel to be supplemented are searched in the temperature-related image to determine the supplement value of the pixel to be supplemented, thus obtaining the constructed region, including: Based on the number of pixels to be added in the constructed region, create a corresponding number of parallel processing threads on the graphics card; Each parallel processing thread determines the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented based on the fast calculation matrix, and determines the supplementary value of the pixel to be supplemented.

5. The method for constructing a three-dimensional temperature map according to claim 4, characterized in that, Each parallel processing thread determines the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented based on the fast calculation matrix, and determines the supplemented value of the pixel to be supplemented, including: For each pixel to be supplemented in the region to be constructed, a preset number of available pixels in its vicinity are determined by the parallel processing thread of the graphics card corresponding to the pixel to be supplemented. Based on a preset number of available pixels near the pixel to be supplemented, the parameters of the three-dimensional cubic interpolation formula corresponding to the pixel to be supplemented are determined in conjunction with the fast calculation matrix. Substitute the parameter and the coordinates of the pixel to be supplemented into the three-dimensional cubic interpolation formula to obtain the supplemented value of the pixel to be supplemented.

6. The method for constructing a three-dimensional temperature map according to claim 5, characterized in that, The preset number of available pixels near the pixel to be supplemented is found based on the Manhattan distance search.

7. A method for constructing a three-dimensional temperature map, characterized in that, include: Acquire temperature-correlation images at several fault locations; wherein at least one of the temperature-correlation images at several fault locations passes through a heating center; For each pixel to be supplemented in the region to be constructed, the supplement value of the pixel to be supplemented is determined by combining the distance between the heated pixels in its vicinity and the heating center, and the constructed region is obtained. A three-dimensional temperature map is generated based on the constructed region.

8. The method for constructing a three-dimensional temperature map according to claim 7, characterized in that, The region to be constructed is determined in the following way: Based on the transformation relationship between the patient's 3D model space and the image space to be constructed, and the target region in the patient's 3D model, the region to be constructed in the image space to be constructed is determined.

9. The method for constructing a three-dimensional temperature map according to claim 7, characterized in that, The region to be constructed is determined in the following way: Pixels with amplitudes greater than a preset amplitude threshold and distances from the heating center less than a preset distance threshold in the temperature-related images of each fault are selected as heated pixels. Using the heating center as the center, determine the area to be constructed that covers each heated pixel.

10. The method for constructing a three-dimensional temperature map according to claim 7, characterized in that, For each pixel to be supplemented in the region to be constructed, the supplementation value of the pixel is determined by combining the distance between the heated pixels in its vicinity and the heating center, thus obtaining the constructed region, including: Based on the number of pixels to be added in the area to be built, create a corresponding number of parallel processing threads for the graphics card; Each parallel processing thread determines the amplitude distance formula corresponding to the pixel to be supplemented based on the distance between the points near the pixel to be supplemented and the heating center, and determines the supplement value of the pixel to be supplemented.

11. The method for constructing a three-dimensional temperature map according to claim 10, characterized in that, Each parallel processing thread determines the amplitude distance formula corresponding to the pixel to be supplemented based on the distance between the points near the pixel to be supplemented and the heating center, and determines the supplement value of the pixel to be supplemented, including: For each pixel to be supplemented in the region to be constructed, a preset number of heated pixels are determined by the parallel processing thread corresponding to the pixel to be supplemented, and the distance between each heated pixel and the heating center is calculated. Based on the distance between each heated pixel and the heating center and the temperature difference between each heated pixel, the least squares method is used to fit the amplitude distance formula corresponding to the pixel to be supplemented. Substitute the distance between the pixel to be supplemented and the heating center into the amplitude distance formula for the pixel to be supplemented to calculate the supplement value of the pixel to be supplemented.

12. The method for constructing a three-dimensional temperature map according to claim 11, characterized in that, The preset number of heated pixels near the pixel to be supplemented is found based on Euclidean distance.

13. The method for constructing a three-dimensional temperature map according to claim 1 or 7, characterized in that, The temperature-related image is a temperature difference map, a phase difference map, or a phase map. Generating a three-dimensional temperature map based on the constructed region includes: The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base value to the size of the image space to be constructed to obtain a three-dimensional temperature-related image; By combining the three-dimensional temperature-related image and the temperature map when it is not heated, a three-dimensional temperature map for the current period is generated.

14. The method for constructing a three-dimensional temperature map according to claim 1 or 7, characterized in that, The temperature-related image is a temperature map, and the step of generating a three-dimensional temperature map based on the constructed region includes: The constructed region is filled into the image space to be constructed, or the area around the constructed region is filled with a base temperature and expanded to the size of the image space to be constructed to obtain a three-dimensional temperature map for the current period.

15. The method for constructing a three-dimensional temperature map according to claim 1 or 7, characterized in that, After generating a three-dimensional temperature map based on the constructed region, the method further includes: performing three-dimensional filtering on the generated three-dimensional temperature map to obtain a smoothed three-dimensional temperature map.

16. The method for constructing a three-dimensional temperature map according to claim 7, characterized in that, A three-dimensional temperature map is generated based on the constructed region, and then the resolution of the three-dimensional temperature map for the current period is improved according to any one of claims 1-6.

17. An ablation system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the three-dimensional temperature map construction method as described in any one of claims 1 to 16.

Images