Image processing method and laser interstitial thermotherapy system

Abstract

The invention provides an image processing method and a laser interstitial thermotherapy system. The method comprises the following steps: determining an optical fiber sleeve region in a magnetic resonance image to be processed; determining a target area adjacent to the optical fiber sleeve area; and judging the temperature of the target area, and determining whether the target area contains a liquid leakage area or not. Based on a laser interstitial thermotherapy scene, in combination with the position characteristics, the signal intensity characteristics and the temperature characteristics of the liquid leakage area found by the inventor, the accuracy of segmenting the liquid leakage area is improved in a targeted manner.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to an image processing method and a laser interstitial hyperthermia system. Background Technology

[0002] Laser interstitial hyperthermia (LITT) is a treatment technique that uses optical fibers to deliver light into the body, causing local biological tissue to coagulate and die upon heating. It achieves the goal of clearing target tissue with minimal invasiveness. During use, optical fibers are typically used in conjunction with fiber optic sheaths to prevent localized overheating.

[0003] The use of laser interstitial hyperthermia combined with magnetic resonance imaging (MRI) technology enables two aspects of monitoring: 1. Generating temperature images of the tissue area through MRI temperature imaging, which helps doctors understand the temperature status and improves safety; 2. Based on temperature imaging, it also combines models (Arrhenius model, CEM43, etc.) to simulate and evaluate the current ablation status of the tissue, generating ablation images to provide information reference for doctors' ablation operations.

[0004] Current ablation monitoring is not comprehensive enough, and unexpected risks may still occur during the ablation process. For example, if the fiber optic sheath ruptures, the cooling medium inside the sheath may enter the tissue, leading to increased intracranial pressure and posing a safety threat. Some existing technologies (such as CN114795449A) use a crack-resistant membrane on the outer wall of the fiber optic sheath and a water-sensitive colored material inside the sheath. When the fiber optic sheath is about to rupture, the water-sensitive colored material in the sheath wall is released into the cooling circulation pipeline, causing a color change. The liquid is then transported outside the body through the circulation pipeline and observed. This solution requires a specially designed sheath, which is costly. Furthermore, judging the leakage area solely based on the signal intensity of magnetic resonance imaging is prone to misjudgment. The signal intensity of the leakage area is similar to that of the ablated area and the fiber optic sheath area, making the judgment of the leakage area susceptible to interference from the ablated area and the fiber optic sheath area.

[0005] In response, this invention provides an image processing method and a laser interstitial hyperthermia system that can achieve more accurate and comprehensive monitoring at a lower cost. Summary of the Invention

[0006] This invention provides an image processing method and a laser interstitial hyperthermia system to address the shortcomings of existing technologies, such as high cost or low accuracy in monitoring leakage.

[0007] In a first aspect, the present invention provides an image processing method, comprising:

[0008] Identify the fiber optic sleeve region in the magnetic resonance image to be processed;

[0009] Determine the target area adjacent to the fiber optic sleeve area;

[0010] The temperature of the target area is assessed to determine whether the target area contains a leakage zone.

[0011] Optionally, determining the fiber optic sleeve region in the magnetic resonance image to be processed includes:

[0012] The magnetic resonance image to be processed is input into the first deep learning model to obtain the fiber optic sleeve region in the magnetic resonance image to be processed.

[0013] Alternatively, the magnetic resonance image to be processed can be registered with the positioning scan image, and the fiber optic sleeve region in the positioning scan image can be mapped to the magnetic resonance image to be processed according to the registration relationship.

[0014] Optionally, determining the target area adjacent to the fiber optic sleeve area includes:

[0015] Several low-signal connectivity regions adjacent to the fiber optic sleeve area are segmented out as the target area.

[0016] Optionally, determining the target area adjacent to the fiber optic sleeve area includes:

[0017] Threshold segmentation or segmentation using a second deep learning model is performed within a preset range around the fiber optic sleeve area to obtain several low-signal connected regions.

[0018] The target region is obtained by removing low-signal connected regions with fewer than a preset threshold number of pixels.

[0019] Furthermore, the magnetic resonance image to be processed is a two-dimensional magnetic resonance image. After obtaining several low-signal connected regions, the method further includes:

[0020] If there are low-signal connectivity domains on both sides of the same segment of the fiber optic sleeve area, then these two low-signal connectivity domains will be merged.

[0021] Optionally, the method further includes a position determination step, the position determination step including:

[0022] For each connected component of the target region, determine whether it is located outside the ablated region;

[0023] If the connected region is located outside the ablated area, then the connected region is determined to be a leakage area.

[0024] Furthermore, the location of the ablated area is obtained in the following manner:

[0025] The ablation region in the ablation map is mapped onto the magnetic resonance image to be processed.

[0026] Optionally, the location of the connected components is determined first, and then the temperature of the remaining connected components that are not yet identified as leakage areas is determined.

[0027] Optionally, the temperature of the connected domain can be determined first, and then the low-temperature area within it can be confirmed as the leakage area by its location.

[0028] Optionally, determining whether the target area contains a leakage zone by judging the temperature of the target area includes:

[0029] Generate the current frame temperature map based on the magnetic resonance image to be processed;

[0030] If a low-temperature zone with a temperature lower than the surrounding temperature is determined in the target area based on the current frame temperature map, then the low-temperature zone is identified as a leakage zone.

[0031] Furthermore, the low-temperature region is screened through the following steps:

[0032] Isotherms are generated in the current frame temperature map. If an isotherm in the target area is closed and the temperature of each isotherm gradually decreases from the outside to the inside, the area defined by the isotherm is taken as the low temperature zone.

[0033] Optionally, the method further includes: determining the shape of the low-temperature zone; if the low-temperature zone is not symmetrical about the fiber optic sleeve area, then further determining that the low-temperature zone belongs to the leakage zone.

[0034] Optionally, the process of determining the shape of the low-temperature region shall be performed by referring to the following steps;

[0035] A first symmetrical region is obtained by performing a symmetrical operation on the first connected region with the axis of the optical fiber sheath region as the axis of symmetry; wherein, the first connected region is any one of the plurality of connected regions;

[0036] Whether the first connected region is symmetrical about the optical fiber sleeve region is determined based on the overlap between the first connected region and the first symmetrical region.

[0037] Further, determining whether the first connected region is symmetrical about the fiber optic sleeve region based on the overlap between the first connected region and the first symmetrical region includes:

[0038] Count the number of pixels in the first connected component, the number of pixels in the first symmetrical region, and the number of overlapping pixels between the first connected component and the first symmetrical region;

[0039] If the ratio of the number of overlapping pixels between the first connected region and the first symmetrical region to the sum of the number of pixels in the first connected region and the number of pixels in the first symmetrical region is greater than a preset ratio threshold, then the first connected region is determined to be symmetrical about the fiber optic sleeve region.

[0040] Optionally, after identifying the leakage area, the system may also include information prompts related to the breakage of the output fiber optic sleeve.

[0041] In a second aspect, the present invention also provides a laser interstitial hyperthermia system, characterized in that it includes: a processing module for executing the image processing method described in any of the preceding claims, providing information guidance for the laser interstitial hyperthermia process.

[0042] The image processing method and laser interstitial hyperthermia system provided by this invention have at least the following beneficial effects:

[0043] 1. It can monitor whether the cooling sleeve is ruptured based on the acquired magnetic resonance images, thus improving safety.

[0044] 2. Based on anomaly recognition using magnetic resonance images, only a data processing step is added to the existing magnetic resonance monitoring system, without the need for additional hardware, thus reducing product costs.

[0045] 3. The signal brightness of the "leakage area" and the "ablation area" on the magnetic resonance image are similar, making it difficult for doctors to determine whether the area is an ablation area or a leakage area. This method, combined with the specific scenario of laser interstitial hyperthermia and the location characteristics, signal intensity characteristics, and temperature characteristics of the leakage area discovered by the inventor, specifically improves the accuracy of segmenting the leakage area.

[0046] 4. In some implementations, the location of each low-signal connected region in the target area is first determined, and the low-signal connected regions outside the ablated area are selected and identified as leakage areas. Then, the temperature determination step is performed on the remaining connected regions, which can more efficiently determine the leakage areas in the magnetic resonance image.

[0047] 5. In some implementations, temperature judgment is first performed on each low-signal connected domain in the target area to filter out low-temperature areas that are lower than the surrounding areas. Then, the location of the low-temperature area is judged again to confirm whether the low-temperature area is a leakage area, which improves the judgment confidence. Attached Figure Description

[0048] 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.

[0049] Figure 1 This is a flowchart illustrating an image processing method provided by the present invention;

[0050] Figure 2 This is a schematic diagram of a magnetic resonance image during the laser interstitial hyperthermia process provided by the present invention;

[0051] Figure 3 This is a schematic diagram illustrating the process of determining whether to merge low-signal connected regions in an image processing method provided by the present invention;

[0052] Figure 4 This is a schematic diagram illustrating the process of determining whether a low-signal connected region is symmetrical about the fiber optic sleeve region in an image processing method provided by the present invention. Detailed Implementation

[0053] 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.

[0054] It should be noted that laser interstitial hyperthermia is an existing surgical procedure, and this invention does not involve any improvement to the operation of laser interstitial hyperthermia. Existing laser interstitial hyperthermia procedures already involve acquiring magnetic resonance imaging (MRI) images. This invention merely processes the acquired MRI images to highlight potential leakage areas, providing information support for the physician.

[0055] The following is combined Figures 1-4 This invention describes an image processing method and a laser interstitial hyperthermia system. Figure 1 This is a flowchart illustrating an image processing method provided by the present invention, as shown below. Figure 1 As shown, the method includes:

[0056] S1. Determine the fiber optic sleeve region in the magnetic resonance image to be processed;

[0057] S2. Determine the target area adjacent to the fiber optic sleeve area;

[0058] S3. Determine the temperature of the target area to determine whether the target area contains a leakage zone.

[0059] Specifically, during laser interstitial hyperthermia, magnetic resonance images are periodically acquired from several or consecutive sections of the patient's tissue to generate temperature and ablation maps. The temperature map displays the tissue temperature at the corresponding location, while the ablation map shows the estimated ablation status of the tissue. This allows doctors to understand the temperature status within the planned ablation area and the currently ablated area in the images.

[0060] This method can process either two-dimensional or three-dimensional magnetic resonance images to identify leakage areas. The magnetic resonance image to be processed in step S1 can reuse the magnetic resonance image used when generating the temperature map and ablation map, without needing to be re-acquired. It should also be noted that this method can be executed multiple times for real-time monitoring, or it can be used to process the magnetic resonance image at a specific moment in a single run, according to user needs.

[0061] Figure 2 This is a schematic diagram of a magnetic resonance imaging (MRI) image. Through image registration, the planned ablation boundary 101 can be marked in MRI image 100. 102 is the fiber optic sleeve region, containing an optical fiber that can output ablation energy at its distal end. Region 103 is the ablated area in the current MRI image, and 104 is the leakage area. (It should be noted that...) Figure 2 (This is not a real magnetic resonance image; it is a schematic diagram provided to facilitate understanding of this solution.)

[0062] Step S1 first identifies the fiber optic sleeve region in the magnetic resonance image to be processed. There are several ways to determine the fiber optic sleeve region, such as through image segmentation or by mapping a fiber optic sleeve region identified in another image to the current frame image through image registration. Step S2 further identifies the target region adjacent to the fiber optic sleeve region. For example, by extending a preset distance outward from the fiber optic sleeve region, the target region is segmented within this distance. It should be noted that since the cooling sleeve region, the cooling medium leaking from the cooling sleeve, and the ablated area of ​​the tissue all appear as low signals on the magnetic resonance image, the segmented target region may include both ablated areas and leakage areas. Step S3 further verifies whether the target region is a leakage area by judging the temperature of the potential bleeding area. Specifically, the temperature of the cooling medium in the fiber optic sleeve is lower than the tissue temperature. When the cooling medium leaks, the temperature map shows that "the temperature of the leakage area is lower than the surrounding temperature." The ablation area is affected by the laser thermal effect, and on the temperature map, it appears as "the temperature of the ablated area is higher than the surrounding temperature". Combining the temperature map with the magnetic resonance amplitude map for auxiliary judgment can more accurately screen out the leakage area.

[0063] In addition, regarding the temperature map in step S3, those skilled in the art, based on knowledge of magnetic resonance imaging, know that magnetic resonance images include amplitude information and phase information, and the change in phase is linearly related to the change in temperature. Therefore, the temperature difference can be obtained by transforming based on the phase difference, and the temperature difference combined with the baseline temperature can generate a temperature map, which can indicate the tissue temperature state at a specific fracture site at a certain moment.

[0064] In this embodiment, steps S1 and S2 determine the target area adjacent to the fiber optic sleeve area (in the magnetic resonance amplitude image) based on the positional relationship between the leakage area and the fiber optic sleeve area. Step S3 combines the differences between the leakage area and the ablated area on the temperature map and the location information to judge the target area, thereby comprehensively improving the accuracy of the leakage area judgment.

[0065] In summary, existing technologies for determining leakage during laser interstitial hyperthermia are costly, and relying solely on the signal intensity of magnetic resonance images to determine the leakage area is prone to misjudgment. Furthermore, the determination of the leakage area is affected by interference from ablated areas and fiber optic sheath areas. This embodiment, based on the laser interstitial hyperthermia scenario, combines the spatial location characteristics, signal intensity characteristics, and temperature characteristics of the leakage area discovered by the applicant to comprehensively improve the accuracy of leakage area determination and reduce product costs.

[0066] Based on the previous embodiment, in one embodiment, S1 includes:

[0067] The magnetic resonance image to be processed is input into the first deep learning model to obtain the fiber optic sleeve region in the magnetic resonance image to be processed.

[0068] Specifically, in this embodiment, a first image segmentation model is pre-trained, capable of segmenting the fiber optic sleeve region in the magnetic resonance image. The first image segmentation model can be a deep neural network (DNN), a convolutional neural network (CNN), a residual network (ResNet), a Transformer model, etc.

[0069] Based on the foregoing embodiments, in another embodiment, S1 includes:

[0070] The magnetic resonance image to be processed is registered with the positioning scan image, and the fiber optic sleeve region in the positioning scan image is mapped to the magnetic resonance image to be processed according to the registration relationship.

[0071] Specifically, before the laser interstitial hyperthermia procedure, after placing the fiber optic cannula and fiber, a large-scale positioning scan is usually performed to determine the actual placement position of the fiber optic cannula. That is, the fiber optic cannula area has been pre-determined in the positioning scan image. In this embodiment, the magnetic resonance image to be processed is registered with the positioning scan image, and the fiber optic cannula area in the positioning scan image is mapped to the magnetic resonance image to be processed according to the registration relationship.

[0072] This embodiment accurately determines the location of the fiber optic sleeve region by registering the magnetic resonance image to be processed with the positioning scan image.

[0073] Based on the foregoing embodiments, in one embodiment, S2 includes:

[0074] S21. Segment the area around the fiber optic sleeve region into several low-signal connected domains as the target region.

[0075] Specifically, "around the fiber optic sleeve area" refers to the area within a preset distance range of the fiber optic sleeve area (excluding the fiber optic sleeve area itself). The cooling medium in the leakage area comes from the fiber optic sleeve, meaning the leakage area is located around the fiber optic sleeve. Step S21 divides the area around the fiber optic sleeve area into several connected regions based on the signal characteristics of the leakage area itself. These connected regions could all be leakage areas.

[0076] Based on the previous embodiment, in one embodiment, S21 includes:

[0077] S211. Perform threshold segmentation or segmentation through a second deep learning model within a preset range around the fiber optic sleeve area to obtain several low-signal connected regions.

[0078] S212. Remove low-signal connected regions with fewer than a preset threshold number of pixels to obtain the target region.

[0079] Specifically, step S211, based on the signal intensity characteristics of the leakage area, segments several low-signal connected regions within a preset range around the fiber optic sleeve area (excluding the fiber optic sleeve area). The segmentation method can be threshold segmentation; the specific threshold is related to the type of tissue being ablated and the type of cooling medium, and can be set empirically, without specific limitations here. The segmentation method can also be based on a deep learning model, such as a deep neural network (DNN), convolutional neural network (CNN), residual network (ResNet), Transformer model, etc. Further, in a preferred embodiment, the second image segmentation model is the same as the first image segmentation model in the above embodiment; that is, the image segmentation model segments both the fiber optic sleeve area in the magnetic resonance image and the low-signal connected regions surrounding the fiber optic sleeve area.

[0080] In addition, in this embodiment, step S212 removes low-signal connected regions with a pixel count lower than a preset threshold, which can eliminate noise interference.

[0081] Based on the previous embodiment, in one embodiment, S1 acquires a two-dimensional magnetic resonance image. After S211, the method further includes: if there are low-signal connectivity regions on both sides of the same segment of the fiber optic sleeve region, then the two low-signal connectivity regions are merged.

[0082] Specifically, whether two low-signal connectivity domains are located on opposite sides of the same segment within the fiber optic sleeve area can be determined in several ways. For example, each low-signal connectivity domain has a contact segment with the fiber optic sleeve area. By comparing the alignment of the contact segments of the low-signal connectivity domains on both sides of the fiber optic sleeve area, it can be determined whether they are located on opposite sides of the same segment. Another example is to determine whether the two connectivity domains are located on opposite sides of the same segment based on the alignment of the midpoints of the two contact segments, referring to... Figure 3 To explain, there are two low-signal connectivity domains 1041 and 1042 on both sides of the fiber optic sleeve area. The midpoint of the contact line between low-signal connectivity domain 1041 and the fiber optic sleeve area is P1, and the midpoint of the contact line between low-signal connectivity domain 1041 and the fiber optic sleeve area is P2. If P1 and P2 differ by no more than 4 pixels in the axial direction of the fiber optic sleeve area (of course, other distance thresholds can also be set, such as 3 pixels, 5 pixels, etc.), then the two low-signal connectivity domains 1041 and 1042 are considered to be located on both sides of the same segment of the fiber optic sleeve area, and the two can be merged into one connectivity domain.

[0083] The method for merging two low-signal connected regions can be flexibly chosen. For example, they can be merged through morphological closing operations, or the contour lines of the two connected regions can be extended and connected by splines to form a closed connected region.

[0084] In this embodiment, the connected domain is integrated through judgment processing, which avoids the same low-signal connected domain being divided into two due to signal obstruction in the fiber optic sleeve area, thus reducing the risk of misjudgment.

[0085] Based on the foregoing embodiments, in one embodiment, the method further includes a position determination step, the position determination step comprising:

[0086] For each connected component of the target region, determine whether it is located outside the ablated region;

[0087] If the connected region is located outside the ablated area, then the connected region is determined to be a leakage area.

[0088] Specifically, still refer to Figure 2 , Figure 2 In the example, two connected components, 103 and 104, are segmented, which together constitute the target region. Additionally, Figure 2The location of the ablated area was determined. For each connected region in the target region, it was determined whether it was located outside the ablated area. If it was located outside the ablated area, it could be directly identified as a leakage area. For example, Figure 2 The connected region 103 coincides with the ablated region, while the connected region 104 is located outside the ablated region. Therefore, the connected region can be identified as the leakage region. This is because the low-signal connected region located within the ablated region could be due to either ablation or leakage, while the low-signal connected region outside the ablated region could only be due to leakage.

[0089] This embodiment efficiently and accurately identifies the leakage area outside the ablation area based on the positional relationship between the connected domain and the ablated area.

[0090] Based on the previous embodiment, in one embodiment, the location of the ablated area is obtained in the following manner:

[0091] The ablation region in the ablation map is mapped onto the magnetic resonance image to be processed.

[0092] Specifically, during laser interstitial hyperthermia, magnetic resonance imaging (MRI) images are continuously acquired. MRI images contain amplitude and phase signals, and the phase change is linearly related to the temperature change. Therefore, a temperature difference map can be obtained by transforming the phase difference. The temperature difference map combined with the baseline temperature generates a temperature map. The temperature map indicates the tissue temperature state at a specific slice at a given time. Furthermore, based on the cumulative effect of temperature over time, the ablation status of the tissue can be determined; that is, an ablation map can be generated from the temperature map. It is understood that all ablated pixels in the ablation map constitute the ablated area. In this embodiment, the ablated area in the ablation map is mapped onto the MRI image to be processed to assist in determining whether low-signal connectivity regions are leakage areas. It is understood that when the patient's position and the scanning parameters of the MRI equipment remain unchanged, the ablation map and the current MRI image to be processed have the same location, and the ablated area in the ablation map can be directly mapped onto the current MRI image to be processed. When the patient's position or the scanning parameters of the MRI machine have changed, it is necessary to obtain a mapping relationship through image registration, and then map the ablated area in the ablation image to the current MRI image to be processed according to the mapping relationship.

[0093] In this embodiment, the ablation map generated during the laser interstitial hyperthermia process is reused to help identify the leakage area, which improves the processing efficiency and reduces the product cost.

[0094] Based on the foregoing embodiments, in one embodiment, the location of the connected domain is first determined, and then the temperature of the remaining connected domains that are not determined to be leakage areas is determined.

[0095] Specifically, after step S2, it is first determined whether each connected region in the target area is located outside the ablated region. If the connected region is located outside the ablated region, it is determined that the connected region is a leakage region. The remaining connected regions are then subjected to temperature determination, and the low-temperature regions in these connected regions are determined as leakage regions.

[0096] For leakage areas outside the ablation zone, the temperature difference between them and the surrounding area is smaller than that between leakage areas within the ablation zone and the surrounding area. In this embodiment, the location can be used to more accurately identify such leakage areas. After that, only the connected domains that are not outside the ablation zone need to be processed, which also improves the processing efficiency.

[0097] In another embodiment, a temperature judgment is first performed on the connected domain to preliminarily determine whether there is a leakage area. Then, the low-temperature area can be confirmed as a leakage area by the location.

[0098] Specifically, in step S3, the temperature of each connected region is determined. After the low-temperature region is screened out, it is further determined whether it is located outside the ablation region. If the low-temperature region is located outside the ablation region, it is confirmed as a leakage region again, which improves the confidence of determining whether the low-temperature region outside the ablation region is a leakage region.

[0099] Based on the foregoing embodiments, in one embodiment, determining whether the target area contains a leakage zone by judging the temperature of the target area includes:

[0100] Generate the current frame temperature map based on the magnetic resonance image to be processed;

[0101] If a low-temperature zone with a temperature lower than the surrounding temperature is determined in the target area based on the current frame temperature map, then the low-temperature zone is identified as a leakage zone.

[0102] Specifically, the temperature of the cooling medium in the fiber optic sleeve is lower than the tissue temperature. When the cooling medium leaks, the temperature map shows that "the temperature of the leaking area is lower than the surrounding temperature." Conversely, the ablation area, affected by the laser thermal effect, shows that "the temperature of the ablated area is higher than the surrounding temperature." It is understood that "surrounding area" here is not limited to the target area; it can be defined by extending a certain distance outward from the target area. Preferably, the "surrounding area" excludes the portion belonging to the fiber optic sleeve area. Temperature comparisons can be made, for example, by comparing statistical average temperatures, or by comparing samples taken from the central area. For instance, a point could be taken in the center of the target area, and another point could be taken in the middle of the ring-shaped area of ​​the "surrounding area," and the temperatures of the two points could be compared.

[0103] This embodiment uses the current frame temperature map to assist in judging whether there is a leakage area in the target area, and more accurately filters out the leakage area.

[0104] Based on the previous embodiment, in one embodiment, the low-temperature region is screened through the following steps:

[0105] Isotherms are generated in the current frame temperature map. If an isotherm L in the target area is closed, and the temperature of each isotherm gradually decreases from the outside to the inside, then the area enclosed by the isotherm L is taken as the low temperature zone.

[0106] Based on the previous embodiment, in one embodiment, the method further includes:

[0107] The shape of the low-temperature zone is determined. If the low-temperature zone is not symmetrical about the fiber optic sleeve area, then it is further determined that the low-temperature zone belongs to the leakage zone.

[0108] Specifically, the inventors discovered that in scenarios where ablation is performed using optical fibers that emit scattered (circumferential) light, the ablated area is usually symmetrical about the fiber optic sleeve area, while the leakage area is usually not symmetrical about the fiber optic sleeve area. This embodiment further determines whether the low-temperature area is symmetrical about the fiber optic sleeve area. If a certain low-temperature area is not symmetrical about the fiber optic sleeve area, it is confirmed to be a leakage area (with a high degree of confidence). If a certain low-temperature area is symmetrical about the fiber optic sleeve area, the user needs to pay further attention, as this area may have temperature anomalies or a low-probability symmetrical leakage area.

[0109] Based on the previous embodiment, in one embodiment, the process of determining the shape of the low-temperature region is performed by referring to the following steps;

[0110] A first symmetrical region is obtained by performing a symmetry operation on the first connected region with the axis of the fiber optic sleeve region as the axis of symmetry; wherein, the first connected region is any one of several connected regions;

[0111] Whether the first connected region is symmetrical about the fiber optic sleeve region is determined based on the overlap between the first connected region and the first symmetrical region.

[0112] Specifically, refer to Figure 4 The first connected region 104 is symmetrical about the optical fiber sleeve region to obtain the first symmetrical region 105. The first connected region 104 and the first symmetrical region 105 are compared to determine whether the first connected region is symmetrical about the optical fiber sleeve region.

[0113] Based on the previous embodiment, in one embodiment, determining whether the first connected region is symmetrical about the fiber optic sleeve region based on the overlap between the first connected region and the first symmetrical region includes:

[0114] Count the number of pixels in the first connected region, the number of pixels in the first symmetrical region, and the number of overlapping pixels between the first connected region and the first symmetrical region;

[0115] If the ratio of the number of overlapping pixels between the first connected region and the first symmetrical region to the sum of the number of pixels in the first connected region and the number of pixels in the first symmetrical region is greater than a preset ratio threshold, then the first connected region is determined to be symmetrical about the fiber optic sleeve region.

[0116] Specifically, still refer to Figure 4 Let S be the number of pixels in the first connected region 104. a Let S be the number of pixels within the first symmetrical region 105. b Let S be the number of pixels that overlap between the two. ab , with indicator 2S ab / (S a +S b As the degree of overlap between the two, when the degree of overlap is greater than the overlap threshold S th At that time, it is determined that the first connected region 104 is symmetrical about the axis of the fiber optic sleeve region. Of course, the above determination method can also be equivalently modified. For example, the number of pixels belonging exclusively to the first connected region 104 and the number of pixels belonging exclusively to the first symmetrical region 105 can be counted separately. The smaller the sum of the two numbers, the higher the degree of overlap, and the better the symmetry of the first connected region 104.

[0117] This embodiment accurately determines whether the low-signal connectivity region relates to the fiber optic sleeve area, effectively eliminating the interference of the ablated area on the determination of the leakage area.

[0118] Based on the foregoing embodiments, in one embodiment, if S3 determines that the target area is a leakage area, then S3 is followed by: S4, prompting information related to the breakage of the output optical fiber sleeve.

[0119] For example, displaying text prompts such as "fiber optic sleeve rupture" on the display screen, outputting audible / visual alarms, providing special displays of the leakage area on the display screen, or prompting to stop the ablation operation, etc.

[0120] The laser interstitial hyperthermia system provided by the present invention is described below. The laser interstitial hyperthermia system described below can be referred to in correspondence with the image processing method described above.

[0121] The present invention provides a laser interstitial hyperthermia system comprising: a processing module, used to execute the image processing method described in any of the preceding claims, to process images acquired during laser interstitial hyperthermia, to determine leakage, and to provide information guidance for the laser interstitial hyperthermia process.

[0122] Optionally, the processing module may implement the steps of the image processing method described above by using a general-purpose computer architecture to run a program. Specifically, the processing module includes 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 image processing method described in any of the preceding claims.

[0123] Optionally, the processing module may also include multiple hardware units, each of which executes one or more steps of the aforementioned image processing method, thereby implementing the aforementioned image processing method as a whole. In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with technological advancements, many improvements to the method flow today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement to the method flow cannot be implemented using a hardware entity module. For example, a Programmable Logic Device (PLD) (e.g., a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used when writing program development code. The original code before compilation must also be written in a specific programming language, which is called a Hardware Description Language (HDL). There is not just one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using the aforementioned hardware description languages ​​and programming it into an integrated circuit, the hardware circuit that implements the logic method flow can be easily obtained.

[0124] The systems described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a computer. A computer can be a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0125] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0126] Those skilled in the art will understand that the descriptions of the various embodiments have different focuses, and parts not described in detail in a certain embodiment can be referred to in the relevant descriptions of other embodiments. Furthermore, it is understood that those skilled in the art, after reading this specification, can conceive of any combination of some or all of the embodiments listed in this specification without creative effort, and such combinations are also within the scope of disclosure and protection of this specification.

[0127] 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. An image processing method, characterized in that, include: Identify the fiber optic sleeve region in the magnetic resonance image to be processed; Determine the target area adjacent to the fiber optic sleeve area; The temperature of the target area is assessed to determine whether the target area contains a leakage zone.

2. The image processing method according to claim 1, characterized in that, The process of determining the fiber optic sleeve region in the magnetic resonance image to be processed includes: The magnetic resonance image to be processed is input into the first deep learning model to obtain the fiber optic sleeve region in the magnetic resonance image to be processed. Alternatively, the magnetic resonance image to be processed can be registered with the positioning scan image, and the fiber optic sleeve region in the positioning scan image can be mapped to the magnetic resonance image to be processed according to the registration relationship.

3. The image processing method according to claim 1, characterized in that, Determining the target area adjacent to the fiber optic sleeve area includes: Several low-signal connectivity regions adjacent to the fiber optic sleeve area are segmented out as the target area.

4. The image processing method according to claim 3, characterized in that, Determining the target area adjacent to the fiber optic sleeve area includes: Threshold segmentation or segmentation using a second deep learning model is performed within a preset range around the fiber optic sleeve area to obtain several low-signal connected regions. The target region is obtained by removing low-signal connected regions with fewer than a preset threshold number of pixels.

5. The image processing method according to claim 3, characterized in that, The method further includes a position determination step, which includes: For each connected component of the target region, determine whether it is located outside the ablated region; If the connected region is located outside the ablated area, then the connected region is determined to be a leakage area.

6. The image processing method according to claim 5, characterized in that, The location of the ablated area was obtained in the following manner: The ablation region in the ablation map is mapped onto the magnetic resonance image to be processed.

7. The image processing method according to claim 5, characterized in that, First, perform a position check on the connected components, and then perform a temperature check on the remaining connected components whose presence of leakage zones is still uncertain.

8. The image processing method according to claim 1, characterized in that, The step of determining the temperature of the target area and whether the target area contains a leakage zone includes: Generate the current frame temperature map based on the magnetic resonance image to be processed; If a low-temperature zone with a temperature lower than the surrounding temperature is determined in the target area based on the current frame temperature map, then the low-temperature zone is identified as a leakage zone.

9. The image processing method according to claim 8, characterized in that, The low-temperature zone is selected through the following steps: Isotherms are generated in the current frame temperature map. If an isotherm in the target area is closed and the temperature of each isotherm gradually decreases from the outside to the inside, the area defined by the isotherm is taken as the low temperature zone.

10. The image processing method according to claim 8, characterized in that, The method also includes: The shape of the low-temperature zone is determined. If the low-temperature zone is not symmetrical about the fiber optic sleeve area, then it is further determined that the low-temperature zone belongs to the leakage zone.

11. The image processing method according to claim 1, characterized in that, After identifying the leakage area, the system also includes information prompts related to the breakage of the output fiber optic sleeve.

12. A laser interstitial hyperthermia system, characterized in that, include: The processing module is used to execute the image processing method according to any one of claims 1-11 to provide information guidance for the laser interstitial hyperthermia process.

Images