Methods, equipment, and storage media for determining ablation regions based on medical images

Abstract

This application discloses a method, device, and storage medium for determining ablation regions based on medical images, belonging to the field of image processing technology. It acquires preoperative images of the lesion tissue in the target area before ablation and postoperative images after ablation; registers and models the preoperative and postoperative images in three dimensions to obtain three-dimensional preoperative and postoperative images; based on the voxel differences of each voxel in the three-dimensional preoperative and postoperative images, it aggregates the voxels to obtain at least one aggregated region; displays the aggregated region on a display interface for the operator to determine whether the aggregated region is an ablation region; if the operator determines that the aggregated region is an ablation region, it displays the ablation region according to a preset display method; this can solve the problem of potentially inaccurate machine recognition of ablation regions and improve the accuracy of ablation region recognition.

Description

Technical Field

[0001] This application belongs to the field of image processing technology, specifically relating to a method, device and storage medium for determining ablation regions based on medical images. Background Technology

[0002] Thermal ablation therapy is a technique that uses thermal effects to coagulate, necrose, vaporize, or carbonize diseased tissue to achieve ablation and inactivation. Thermal ablation techniques include laser therapy, high-frequency electrosurgical excision, argon plasma coagulation, microwave therapy, and radiofrequency ablation. Taking Laser Interstitial Thermal Therapy (LITT) as an example, LITT is a novel minimally invasive technique for various intracranial lesions. It has been used to treat deep, difficult-to-resecte intracranial lesions, such as tumors, radiation necrosis, and epileptic foci in the brain's deep nuclei and white matter. This technique uses a laser-tipped probe inserted into the brain lesion, generating controlled thermal damage by heating surrounding tissue. Intraoperative real-time magnetic resonance imaging (MRI) temperature measurement allows for continuous monitoring of the ablation area, and ablation can be stopped at any time.

[0003] To ensure the effectiveness of thermal ablation therapy, it is necessary to rely on modern medical imaging technology to accurately assess the ablation status of the lesion tissue after treatment. Image recognition technology plays a crucial role in the postoperative efficacy evaluation of laser ablation therapy. How to quickly and accurately identify the postoperative ablation area is an important prerequisite for postoperative evaluation. Currently, existing ablation area identification methods mainly include deep learning and manual delineation. Specifically: ① Deep learning involves inputting preoperative and postoperative images into a deep learning network model to identify the ablation area. The deep learning network model is pre-trained using a dataset. The dataset includes sample preoperative images, corresponding sample postoperative images, and the labeled ablation area in the sample postoperative image. However, a drawback of deep learning is that it requires a large number of samples for training to achieve good generalization. Currently, the amount of dataset for certain areas (such as the brain) may be limited, resulting in insufficient datasets for training the deep learning network model. Furthermore, deep learning-based image recognition technology cannot quickly and accurately identify the ablation area; when encountering artifacts or noise, this identification method still struggles to achieve ideal results. ② Manually delineated ablation area identification methods are time-consuming and highly dependent on the clinician's clinical experience, thus having certain limitations. ③ Furthermore, for situations requiring multiple tumors to be ablated at multiple locations using multiple optical fibers, there is currently no effective identification method that can achieve rapid and accurate identification of multiple tumor areas or multiple ablation areas in a single operation.

[0004] Existing methods for identifying ablation regions are lacking in speed and accuracy, failing to meet practical needs. Therefore, how to quickly and accurately identify ablation regions is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by this application includes the issue that the identification results may be inaccurate when the machine identifies the ablation area.

[0006] To address the aforementioned technical problems, this application provides a method for determining ablation regions based on medical images, the method comprising:

[0007] Acquire preoperative images of the lesion tissue in the target area before ablation, and postoperative images after ablation;

[0008] The preoperative and postoperative images are registered and 3D modeled to obtain 3D preoperative and 3D postoperative images.

[0009] Based on the voxel differences of each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image, the voxels are aggregated to obtain at least one aggregated region.

[0010] The aggregation area is displayed on the screen so that the operator can determine whether the aggregation area is an ablation area;

[0011] If the operator determines that the aggregation area is an ablation area, the ablation area is displayed according to a preset display method.

[0012] Optionally, the step of performing region aggregation on each voxel based on the voxel difference values ​​in the three-dimensional preoperative image and the three-dimensional postoperative image to obtain at least one aggregated region includes:

[0013] Calculate the voxel differences of each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image;

[0014] Obtain the region aggregation threshold;

[0015] For the target voxel that has not been aggregated in each voxel, if the voxel difference between the first adjacent voxels of the target voxel is less than the region aggregation threshold, the first adjacent voxel and the target voxel are aggregated into the same region, and it is determined whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold.

[0016] If the voxel difference between the second adjacent voxels is less than the region aggregation threshold, the second adjacent voxel and the target voxel are aggregated into the same region, and the second adjacent voxel is used as the first adjacent voxel. The step of determining whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold is executed again to obtain the aggregation region corresponding to the target voxel.

[0017] Optionally, obtaining the regional aggregation threshold includes:

[0018] A voxel difference list is generated based on the voxel differences of each voxel; the voxel difference list includes the voxel difference corresponding to each voxel, and each voxel difference corresponds to the position information of at least one voxel; the voxel differences in the voxel difference list are arranged from largest to smallest.

[0019] The region aggregation threshold is obtained by calculating the difference between the target voxel difference corresponding to the target voxel and the difference between the nth voxel difference value in the voxel difference list after the target voxel difference value; where n is a positive integer.

[0020] Optionally, the method further includes:

[0021] The target point for ablation of lesions is determined, and the target voxel is obtained. The target point refers to the set of spatial coordinates of the optical fiber output when ablation is performed on all lesions.

[0022] Optionally, the method further includes:

[0023] The target voxel is obtained by identifying the unpolymerized voxel with the largest voxel difference from the individual voxels.

[0024] Optionally, displaying the aggregation area on the display interface includes:

[0025] Determine whether the maximum edge distance of the aggregation region is greater than a preset threshold, the preset threshold being determined based on the ablation diameter;

[0026] If the maximum edge distance is greater than the preset threshold, the aggregated area is displayed on the display interface.

[0027] Optionally, when the operator determines that the aggregation area is an ablation area, displaying the ablation area according to a preset display method includes:

[0028] In response to receiving a confirmation input from the operator, the ablation area is displayed according to a preset display method.

[0029] Optionally, if the operator determines that the aggregation area is an ablation area, displaying the ablation area according to a preset display method includes:

[0030] A three-dimensional model of the ablation region is performed to obtain the three-dimensional ablation region;

[0031] The three-dimensional ablation area is displayed according to the preset display method.

[0032] On the other hand, this application also provides an electronic device, which includes a processor and a memory; the memory stores a program, which is loaded and executed by the processor to implement the above-described method for determining ablation regions based on medical images.

[0033] On the other hand, this application also provides a computer-readable storage medium storing a program that, when executed by a processor, is used to implement the above-described method for determining ablation regions based on medical images.

[0034] The technical solution provided in this application has the following advantages:

[0035] By acquiring preoperative images of the lesion tissue in the target area before ablation and postoperative images after ablation; registering and 3D modeling the preoperative and postoperative images to obtain 3D preoperative and 3D postoperative images; based on the voxel differences of each voxel in the 3D preoperative and 3D postoperative images, region aggregation is performed on each voxel to obtain at least one aggregated region; the aggregated region is displayed on the display interface for the operator to determine whether the aggregated region is the ablation region; if the operator determines that the aggregated region is the ablation region, the ablation region is displayed according to the preset display method; this can solve the problem that the recognition result may be inaccurate when the machine identifies the ablation region; by performing region aggregation on each voxel and pushing the obtained aggregated region to the operator for the judgment of the ablation region, the accuracy of ablation region identification can be improved.

[0036] In addition, by determining the region aggregation threshold based on the position of the target voxel difference in the voxel difference list, the region aggregation threshold can be dynamically adjusted adaptively, which can improve the accuracy of region aggregation.

[0037] In addition, since voxels with large voxel differences are more likely to be ablation regions, by prioritizing the aggregation of voxels with large voxel differences in descending order and recommending them first, ablation regions can be recommended as much as possible, thus improving the region recommendation effect.

[0038] In addition, since there is a high probability that the target site has an ablation region, prioritizing the recommendation of the target site's ablation region can further improve the region recommendation effect.

[0039] Furthermore, since the purpose of pushing the aggregation area is to allow the operator to select the ablation region, and the size of the ablation region often meets preset size conditions, displaying the aggregation area on the display interface when the maximum edge distance is greater than a preset threshold can reduce the number of aggregation areas the operator needs to judge, thereby improving judgment efficiency.

[0040] Furthermore, for cases where multiple tumors require ablation at multiple locations using multiple optical fibers, the present invention displays the aggregated area on the display interface when the maximum edge distance exceeds a preset threshold. By using the identification method of the present invention, the expected ablation area can be identified in a one-time quantifiable manner or recommended to medical staff first, and then the medical staff can confirm whether the recommended area is an ablation area, thus avoiding misjudgment. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0042] Figure 1 This is a flowchart of a method for determining ablation regions based on medical images, provided in one embodiment of this application;

[0043] Figure 2 This is a block diagram of an ablation region determination device based on medical images provided in one embodiment of this application;

[0044] Figure 3 This is a block diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0045] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. The application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0047] In this application, unless otherwise stated, directional terms such as "upper," "lower," "top," and "bottom" are generally used in relation to the direction shown in the accompanying drawings, or in relation to the vertical, perpendicular, or gravitational direction of the component itself; similarly, for ease of understanding and description, "inner" and "outer" refer to the inner and outer contours of each component itself, but the above directional terms are not intended to limit this application.

[0048] In this application, the implementation subject of each embodiment is an electronic device as an example. The electronic device has image processing function. The electronic device can be a computer, tablet computer, medical imaging device, etc. This embodiment does not limit the implementation method of the electronic device.

[0049] Figure 1 This is a flowchart of a method for determining ablation regions based on medical images according to an embodiment of this application. The method includes at least the following steps:

[0050] Step 101: Obtain preoperative images of the lesion tissue in the target area before ablation and postoperative images after ablation.

[0051] Optionally, preoperative and postoperative images include magnetic resonance imaging (MRI) images. In other embodiments, preoperative and postoperative images may also include computed tomography (CT) images; this embodiment does not limit the image types of preoperative and postoperative images.

[0052] Optionally, the preoperative and postoperative images can be a frame from a video stream or a single image. This embodiment does not limit the source of the preoperative and postoperative images.

[0053] Optionally, the preoperative and postoperative images can be acquired by electronic devices or sent by other devices that are connected to the electronic devices. This embodiment does not limit the way the electronic devices acquire the preoperative and postoperative images.

[0054] Step 102: Register and 3D model the preoperative and postoperative images to obtain 3D preoperative and 3D postoperative images.

[0055] Optionally, the electronic device registers the preoperative and postoperative images, and then performs 3D modeling on the registered images respectively; or, the electronic device performs 3D modeling on the preoperative and postoperative images, and then registers the 3D modeled images. This embodiment does not limit the execution order between registration and 3D modeling.

[0056] Image registration methods include using point set registration techniques. These point set registration techniques include, but are not limited to, Iterative Closest Point (ICP) algorithms, Robust Point Matching (RPM) algorithms, Kernel Correlation (KC) algorithms, or Coherent Point Drift (CPD) algorithms. This embodiment does not limit the image registration method.

[0057] Three-dimensional modeling of images can be performed using either surface-based reconstruction algorithms or volume-based reconstruction algorithms. This embodiment does not limit the method of three-dimensional modeling.

[0058] Step 103: Based on the voxel differences of each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image, perform region aggregation on each voxel to obtain at least one aggregated region.

[0059] Since the voxel difference in the ablation region is large and the voxel difference in the non-ablation region is small, in this embodiment, the ablation region and the non-ablation region can be distinguished by the voxel difference, so that the operator can select the ablation region from the various aggregated regions.

[0060] In one example, based on the voxel differences between individual voxels in the 3D preoperative and 3D postoperative images, region aggregation is performed on each voxel to obtain at least one aggregated region, including the following steps:

[0061] Step 1: Calculate the voxel differences of each voxel in the 3D preoperative and 3D postoperative images.

[0062] To illustrate, if the grayscale value ranges from 0 to 255, the voxel difference can be any value from 0 to 255. Alternatively, if the grayscale value ranges from 0 to 65535, the voxel difference can be any value from 0 to 65535.

[0063] Step 2: Obtain the region aggregation threshold.

[0064] Optionally, obtaining the region aggregation threshold includes: generating a voxel difference list based on the voxel differences of each voxel; for each unaggregated target voxel, calculating the difference between the target voxel difference corresponding to the target voxel and the nth voxel difference in the voxel difference list after the target voxel difference, to obtain the region aggregation threshold. n is a positive integer.

[0065] The voxel difference list includes the voxel difference for each voxel, and each voxel difference corresponds to the position information of at least one voxel; the voxel differences in the voxel difference list are arranged from largest to smallest.

[0066] For example, the list of voxel differences is shown in Table 1 below. According to Table 1, the voxel differences are arranged in descending order, and each voxel difference corresponds to the position information of at least one voxel.

[0067] Table 1:

[0068]

[0069]

[0070] Optionally, the target voxel is the target location during lesion ablation. In this case, before step 2, it is necessary to determine the target location during ablation to obtain the target voxel. The target location can be one, at least two, or more. The target location is the position where the ablation energy source acts. Taking LITT as an example, the target location is the position where the optical fiber acts. More specifically, the target point is the ablation location where the light-emitting part of the optical fiber undergoes ablation within the brain. Preferably, the target point is the set of spatial coordinates of the light emitted by the optical fiber during ablation of all lesion tissues. Furthermore, the target point can also be the center of the light-emitting part of the optical fiber or the end of the optical fiber. After laser ablation, the ablation area at the target location can be determined preferentially or only aggregated based on the target location information and the voxel difference matrix, accelerating the speed and accuracy of ablation area identification.

[0071] Alternatively, the target voxel can be the unpolymerized voxel with the largest voxel difference among all voxels. In this case, before step 2, it is necessary to determine the unpolymerized voxel with the largest voxel difference from all voxels to obtain the target voxel.

[0072] In actual implementation, the electronic device can first determine the aggregation region corresponding to the target point, and then determine the aggregation region corresponding to the voxel with the largest voxel difference. This embodiment does not limit the order of division of aggregation regions.

[0073] Optionally, the value of n can be a fixed value, or determined based on the range of voxel differences. When n is determined based on the range of voxel differences, the value of n is the product of the range of voxel differences and a preset percentage.

[0074] Furthermore, if the preset percentage is too small, the ablation area recommendation will be incomplete; if it is too large, the system will recommend too many ablation areas, resulting in inaccurate identification. Therefore, setting an appropriate preset percentage is extremely important. The preset percentage can also be adaptively adjusted according to the MRI equipment model, operating parameters, and / or operating environment, so that the ablation area determination method provided in this application is applicable to any MRI equipment and avoids the correction process of the image data itself, thereby accelerating the speed of ablation area identification.

[0075] In another embodiment of this application, the preset percentage can be 10%, 15%, or any value between 10% and 15%, or other values ​​set by the operator, such as 5% or 21%. This embodiment does not limit the value of the preset percentage.

[0076] Suppose that the target voxel V1 is represented by voxel X1 in the voxel difference list, and the next 10% of voxels in the same voxel difference list are represented by voxels X2. Then, the region aggregation threshold DeltaX = V(X2) - V(X1). Here, V(X2) and V(X1) represent the voxel differences between the two voxels.

[0077] In other embodiments, the region aggregation threshold may also be a preset fixed value. This embodiment does not limit the way the region aggregation threshold is set.

[0078] Step 3: For the target voxel that has not been aggregated in each voxel, if the voxel difference between the first adjacent voxels of the target voxel is less than the region aggregation threshold, aggregate the first adjacent voxel and the target voxel into the same region, and determine whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold.

[0079] Optionally, if the voxel difference between the first adjacent voxels of the target voxel is greater than or equal to the region aggregation threshold, it is determined that the first adjacent voxel and the target voxel do not belong to the same aggregation region. If the voxel difference between the second adjacent voxels is greater than or equal to the region aggregation threshold, it is determined that the second adjacent voxel and the target voxel do not belong to the same aggregation region. Here, the first adjacent voxel and the second adjacent voxel are both any voxels surrounding or near the target voxel. Further, the voxel difference of the target voxel is greater than the voxel difference of the first adjacent voxel, and the voxel difference of the first adjacent voxel is greater than the voxel difference of the second adjacent voxel; the selection or definition of the first and second adjacent voxels is not limited here.

[0080] Step 4: If the voxel difference between the second adjacent voxels is less than the region aggregation threshold, the second adjacent voxel and the target voxel are aggregated into the same region, and the second adjacent voxel is taken as the first adjacent voxel. The step of determining whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold is executed again to obtain the aggregated region corresponding to the target voxel.

[0081] In other examples, the method of region aggregation based on the voxel difference of each voxel to obtain at least one aggregated region can also be: determine the voxel whose voxel difference changes abruptly, use the voxel as the edge of the aggregated region, and obtain at least one aggregated region. This embodiment does not limit the method of region aggregation.

[0082] Among them, a sudden change in voxel difference means that the difference between two adjacent voxels is greater than a preset threshold.

[0083] Step 104: Display the aggregation area on the display interface so that the operator can determine whether the aggregation area is an ablation area.

[0084] In one example, the electronic device displays all the aggregated areas on the display interface.

[0085] In another example, since the purpose of pushing aggregated areas is to allow the operator to select ablation regions, and the size of ablation regions often meets preset size conditions, to avoid the problem of low efficiency caused by the operator having to judge a large number of aggregated areas when all aggregated areas are pushed to the operator, the electronic device in this example can determine whether the size of an aggregated area meets the preset size conditions before displaying it on the display interface; if the preset size conditions are met, the aggregated area is displayed on the display interface; if the preset size conditions are not met, the aggregated area is determined not to be an ablation region and is not displayed on the display interface for the operator to judge. This reduces the number of aggregated areas the operator needs to judge, thereby improving judgment efficiency.

[0086] In addition, most noise in the image does not aggregate. Therefore, in the aggregation region determination stage of this application, the image generated by noise can be removed effectively. Although artifacts will aggregate, the image displayed by the artifacts is significantly different from the image of the ablation region, which can be easily identified by the operator. Therefore, the image displayed by the artifacts can be removed.

[0087] Schematic, the preset size conditions include, but are not limited to: the maximum edge distance of the aggregation region is greater than a preset threshold; and / or, the area is greater than a preset area threshold; and / or, the volume is greater than a preset volume threshold.

[0088] In this embodiment, the example given is that the maximum edge distance of the aggregation region is greater than a preset threshold. Displaying the aggregation region on the display interface includes: determining whether the maximum edge distance of the aggregation region is greater than the preset threshold, where the preset threshold is determined based on the ablation diameter; and displaying the aggregation region on the display interface when the maximum edge distance is greater than the preset threshold.

[0089] The preset threshold is either equal to or smaller than the ablation diameter. Furthermore, the preset threshold can be set based on the tumor size and / or the intraoperative ablation monitoring system. The preset threshold is generally a value smaller than the ablation diameter; for example, if the tumor diameter is 2 cm, the preset threshold could be 1 cm.

[0090] Optionally, after determining the aggregation area to be displayed, the aggregation area can be displayed in one of the following ways, including but not limited to:

[0091] The first method involves displaying all aggregated areas at once, and then having the operator confirm which of these aggregated areas are ablation areas.

[0092] The second method involves displaying x aggregated regions each time, with the operator confirming whether the currently displayed aggregated region is an ablation region, and then proceeding to step 105; determining whether there are any undisplayed aggregated regions; if so, repeating the step of displaying x aggregated regions each time and having the operator confirm whether the currently displayed aggregated region is an ablation region, until there are no undisplayed aggregated regions left to stop.

[0093] Where x is 1 or an integer greater than 1, this embodiment does not limit the value of x.

[0094] It should be noted that if there are undisplayed aggregated areas and the number of undisplayed aggregated areas is less than x, all remaining undisplayed aggregated areas can be displayed when the last aggregated area is displayed. That is, the number of aggregated areas displayed in the last time can be less than x.

[0095] Step 105: If the operator determines that the aggregation area is the ablation area, the ablation area is displayed according to the preset display method.

[0096] Schematic, when the operator determines that the aggregation area is the ablation area, the ablation area is displayed according to a preset display method, including: in response to receiving a confirmation operation input by the operator, the ablation area is displayed according to the preset display method.

[0097] Optionally, the operator may input confirmation information in ways including but not limited to using a mouse, keyboard, or touchscreen. This embodiment does not limit the input method for confirmation.

[0098] Optionally, the preset display method includes, but is not limited to: highlighting, displaying in a preset color, or displaying with thickened edges. This embodiment does not limit the implementation of the preset display method.

[0099] Since the ablation area needs to be used by the operator to judge the ablation effect, in order to improve the display effect of the ablation area, in this embodiment, when the operator determines that the aggregation area is the ablation area, the ablation area is displayed according to a preset display method, including: performing three-dimensional modeling on the ablation area to obtain a three-dimensional ablation area; and displaying the three-dimensional ablation area according to the preset display method.

[0100] In summary, the ablation region determination method based on medical images provided in this embodiment acquires preoperative images of the lesion tissue in the target area before ablation and postoperative images after ablation; registers and models the preoperative and postoperative images to obtain three-dimensional preoperative and postoperative images; aggregates each voxel based on the voxel difference in the three-dimensional preoperative and postoperative images to obtain at least one aggregated region; displays the aggregated region on the display interface for the operator to determine whether the aggregated region is an ablation region; if the operator determines that the aggregated region is an ablation region, the ablation region is displayed according to a preset display method; this can solve the problem of potentially inaccurate recognition results when the machine identifies the ablation region; by aggregating each voxel and pushing the obtained aggregated region to the operator for ablation region judgment, the accuracy of ablation region identification can be improved.

[0101] In addition, by determining the region aggregation threshold based on the position of the target voxel difference in the voxel difference list, the region aggregation threshold can be dynamically adjusted adaptively, which can improve the accuracy of region aggregation.

[0102] In addition, since voxels with large voxel differences are more likely to be ablation regions, by prioritizing the aggregation of voxels with large voxel differences in descending order and recommending them first, ablation regions can be recommended as much as possible, thus improving the region recommendation effect.

[0103] In addition, since there is a high probability that the target site has an ablation region, prioritizing the recommendation of the target site's ablation region can further improve the region recommendation effect.

[0104] Furthermore, since the purpose of pushing the aggregation area is to allow the operator to select the ablation region, and the size of the ablation region often meets preset size conditions, displaying the aggregation area on the display interface when the maximum edge distance is greater than a preset threshold can reduce the number of aggregation areas the operator needs to judge, thereby improving judgment efficiency.

[0105] Furthermore, for cases where multiple tumors require ablation at multiple locations using multiple optical fibers, the present invention displays the aggregated area on the display interface when the maximum edge distance exceeds a preset threshold. By using the identification method of the present invention, the expected ablation area can be identified in a one-time quantifiable manner or recommended to medical staff first, and then the medical staff can confirm whether the recommended area is an ablation area, thus avoiding misjudgment.

[0106] Figure 2 This is a block diagram of an ablation region determination device based on medical images according to an embodiment of this application. The device includes at least the following modules: an image acquisition module 210, an image processing module 220, a region aggregation module 230, a first display module 240, and a second display module 250.

[0107] The image acquisition module 210 is used to acquire preoperative images of the lesion tissue in the target area before ablation and postoperative images after ablation.

[0108] Image processing module 220 is used to register and 3D model the preoperative image and the postoperative image to obtain 3D preoperative image and 3D postoperative image.

[0109] The region aggregation module 230 is used to perform region aggregation on each voxel based on the voxel difference value on each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image to obtain at least one aggregated region.

[0110] The first display module 240 is used to display the aggregation area on the display interface so that the operator can determine whether the aggregation area is an ablation area;

[0111] The second display module 250 is used to display the ablation area according to a preset display method when the operator determines that the aggregation area is the ablation area.

[0112] For relevant details, please refer to the above method implementation examples.

[0113] It should be noted that the ablation region determination device based on medical images provided in the above embodiments is only illustrated by the division of the above functional modules when determining the ablation region based on medical images. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the ablation region determination device based on medical images can be divided into different functional modules to complete all or part of the functions described above. In addition, the ablation region determination device based on medical images provided in the above embodiments and the ablation region determination method embodiments based on medical images belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0114] Figure 3This is a block diagram of an electronic device provided in one embodiment of this application. The device may be... Figure 1 The electronic device includes at least a processor 301 and a memory 302.

[0115] Processor 301 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 301 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 301 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 301 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 301 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0116] The memory 302 may include one or more computer-readable storage media, which may be non-transitory. The memory 302 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 302 are used to store at least one instruction, which is executed by the processor 301 to implement the ablation region determination method based on medical images provided in the method embodiments of this application.

[0117] In some embodiments, the external parameter calibration device may also optionally include: a peripheral device interface and at least one peripheral device. The processor 301, memory 302, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Indicatively, peripheral devices include, but are not limited to: radio frequency circuits, touch displays, audio circuits, and power supplies.

[0118] Of course, the external parameter calibration device may include fewer or more components, and this embodiment does not limit this.

[0119] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the ablation region determination method based on medical images in the above-described method embodiments.

[0120] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the ablation region determination method based on medical images described in the above method embodiments.

[0121] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0123] Obviously, the embodiments described above are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, those skilled in the art can make other variations or modifications without creative effort, and all such variations or modifications should fall within the scope of protection of this application.

Claims

1. A method for determining ablation regions based on medical images, characterized in that, The method includes: Acquire preoperative images of the lesion tissue in the target area before ablation, and postoperative images after ablation; The preoperative and postoperative images are registered and 3D modeled to obtain 3D preoperative and 3D postoperative images. Based on the voxel differences of each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image, the voxels are aggregated to obtain at least one aggregated region, wherein each voxel difference corresponds to the position information of at least one voxel. The aggregation area is displayed on the screen so that the operator can determine whether the aggregation area is an ablation area; If the operator determines that the aggregation area is an ablation area, the ablation area is displayed according to a preset display method; Based on the voxel differences between each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image, each voxel is region-aggregated to obtain at least one aggregated region, including: Calculate the voxel differences of each voxel in the three-dimensional preoperative image and the three-dimensional postoperative image; obtain the region aggregation threshold; For the target voxel that has not been aggregated in each voxel, if the voxel difference between the first adjacent voxels of the target voxel is less than the region aggregation threshold, the first adjacent voxel and the target voxel are aggregated into the same region, and it is determined whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold. If the voxel difference between the second adjacent voxels is less than the region aggregation threshold, the second adjacent voxel and the target voxel are aggregated into the same region, and the second adjacent voxel is used as the first adjacent voxel. The step of determining whether the voxel difference between the second adjacent voxels of the first adjacent voxel is less than the region aggregation threshold is executed again to obtain the aggregation region corresponding to the target voxel; and Identify the voxel whose voxel difference value changes abruptly, and use that voxel as the edge of the aggregation region. A voxel difference change abruptly means that the difference between the voxel differences of two adjacent voxels is greater than a preset threshold.

2. The method according to claim 1, characterized in that, Obtaining the region aggregation threshold includes: A voxel difference list is generated based on the voxel differences of each voxel; the voxel difference list includes the voxel difference corresponding to each voxel, and each voxel difference corresponds to the position information of at least one voxel; the voxel differences in the voxel difference list are arranged from largest to smallest. The region aggregation threshold is obtained by calculating the difference between the target voxel difference corresponding to the target voxel and the difference between the nth voxel difference value in the voxel difference list after the target voxel difference value; where n is a positive integer.

3. The method according to claim 1, characterized in that, The method further includes: The target point for ablation of lesions is determined, and the target voxel is obtained. The target point refers to the set of spatial coordinates of the optical fiber output when ablation is performed on all lesions.

4. The method according to claim 1, characterized in that, The method further includes: The target voxel is obtained by identifying the unpolymerized voxel with the largest voxel difference from the individual voxels.

5. The method according to claim 1, characterized in that, The aggregation area is displayed on the display interface, including: Determine whether the maximum edge distance of the aggregation region is greater than a preset threshold, the preset threshold being determined based on the ablation diameter; If the maximum edge distance is greater than the preset threshold, the aggregated area is displayed on the display interface.

6. The method according to claim 1, characterized in that, When the operator determines that the aggregation area is an ablation area, displaying the ablation area according to a preset display method includes: In response to receiving a confirmation input from the operator, the ablation area is displayed according to a preset display method.

7. The method according to any one of claims 1 to 6, characterized in that, When the operator determines that the aggregation area is an ablation area, displaying the ablation area according to a preset display method includes: A three-dimensional model of the ablation region is performed to obtain the three-dimensional ablation region; The three-dimensional ablation area is displayed according to the preset display method.

8. An electronic device, characterized in that, The device includes a processor and a memory; the memory stores a program that is loaded and executed by the processor to implement the ablation region determination method based on medical images as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The storage medium stores a program that, when executed by a processor, is used to implement the ablation region determination method based on medical images as described in any one of claims 1 to 7.

Images