Image registration method, system, device and medium
By using Histoimage for image segmentation and registration in PET image registration, the problems of poor real-time PET image registration and large calculation overhead in the prior art are solved, and efficient and real-time image registration is achieved.
Patent Information
- Application Number
- CN202310955662.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-31
- Publication Date
- 2025-02-14
AI Technical Summary
The existing PET image registration process has poor real-time performance and high computational overhead, which limits its application in time-sensitive scenarios.
Using the Histoimage-based image registration method, by acquiring the PET modal Histoimage, image segmentation and registration processing are performed, and deformation fields are generated, replacing the use of traditional reconstructed images.
It improves the real-time and efficiency of PET image registration, reduces calculation overhead, and ensures registration accuracy.
Smart Images
Figure CN119444811A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of image processing technology, and in particular to an image registration method, system, device and medium. Background Art
[0002] Image registration is a basic tool in medical image analysis and is widely used in modality fusion, pathological analysis and diagnosis, computer-assisted surgery and other fields. Image registration usually involves two images, namely the reference image (fixed image) and the moving image (image to be registered). Image registration is the process of spatially matching two images of different or same modalities. The result is the spatial transformation relationship of the moving image to the reference image. The goal of registration is to obtain a deformation field or other parameters that describe relative motion, so as to determine the corresponding position of each voxel in the reference image in the moving image.
[0003] Images of different modalities refer to images obtained using different imaging principles and different imaging devices, such as CT (computed tomography) images, MRI (nuclear magnetic resonance imaging) images, and PET (positron emission tomography) images, which are all images of different modalities.
[0004] In the existing PET modality image registration, in order to obtain a PET image with good anatomical structure information, it is necessary to first obtain a reconstructed image. Specifically, the directly acquired PET raw data is processed into a reconstructed image through methods such as FBP (Filtered backprojection) or OSEM (Ordered Subsets Expectation Maximization). After obtaining the reconstructed image, the registration process of the motion image and the reference image is performed. Since the image reconstruction process leads to excessive time and computational overhead, the application of this technology in time-sensitive scenarios is limited. After obtaining the reconstructed image, the time spent on the registration itself is usually less than the reconstruction time. The excessive reconstruction time and computational threshold reduce the practicality of image registration. Summary of the invention
[0005] The technical problem to be solved by the present disclosure is to overcome the defects of poor real-time performance and high computational overhead in the PET image registration process in the prior art, and to provide an image registration method, system, device and medium.
[0006] The present invention solves the above technical problems through the following technical solutions:
[0007] In a first aspect, a method for image registration based on Histoimage (back projection image) is provided, the method comprising:
[0008] Acquire the image to be registered and the reference image of the target object;
[0009] Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality;
[0010] The image to be registered and the reference image are subjected to image segmentation processing, and the image to be registered and the reference image after the segmentation processing are subjected to image registration processing to generate a deformation field for registering the image to be registered to the reference image, so as to realize image registration of the image to be registered.
[0011] Preferably, the step of acquiring the image to be registered and the reference image of the target object comprises:
[0012] Acquire several groups of raw data corresponding to the target object during the PET scanning process;
[0013] Generate a plurality of target Histoimages corresponding to the target object based on a plurality of groups of raw data;
[0014] Using different target Histoimages as the image to be registered and the reference image;
[0015] Or, any of the target Histoimages is used as the image to be registered or the reference image.
[0016] Preferably, the step of generating a plurality of target Histoimages corresponding to the target object based on the plurality of sets of raw data comprises:
[0017] Based on each group of raw data, generate a number of initial Histoimages corresponding to the target object;
[0018] The initial Histoimage is corrected to generate the corresponding target Histoimage.
[0019] Preferably, the step of generating an initial Histoimage corresponding to the target object based on each set of raw data comprises:
[0020] Obtaining annihilation events in each set of the birth data;
[0021] Performing a back-projection operation on the annihilation event to generate the corresponding initial Histoimage;
[0022] And / or, the correction operation includes at least one of scatter correction, attenuation correction, sensitivity correction and random correction.
[0023] Preferably, the step of performing image segmentation processing on the image to be registered and the reference image, performing image registration processing on the segmented image to be registered and the reference image, and generating a deformation field for registering the image to be registered to the reference image comprises:
[0024] Based on a preset segmentation algorithm, performing image segmentation processing on the image to be registered and the reference image to obtain a first image corresponding to the image to be registered and a second image corresponding to the reference image;
[0025] Based on a preset registration algorithm, the first image and the second image are matched to generate the deformation field for registering the image to be registered to the reference image.
[0026] Preferably, when only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other one is different from the PET modality.
[0027] Preferably, when only the image to be registered is the Histoimage of the PET modality, the reference image includes a CT modality image or an MRI modality image;
[0028] Or, when the only reference image is a Histoimage in the PET modality, the image to be registered includes a CT modality image or an MRI modality image.
[0029] In a second aspect, a Histoimage-based image registration system is also provided, the image registration system comprising:
[0030] An image acquisition module, used to acquire the image to be registered and the reference image of the target object;
[0031] Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality;
[0032] The image registration module is used to perform image segmentation processing on the image to be registered and the reference image, and perform image registration processing on the image to be registered and the reference image after the segmentation processing, and generate a deformation field for registering the image to be registered to the reference image, so as to realize image registration of the image to be registered.
[0033] Preferably, the image acquisition module comprises:
[0034] A raw data acquisition unit, used to acquire several groups of raw data corresponding to the target object during the PET scanning process;
[0035] A target Histoimage generating unit, used for generating a plurality of target Histoimages corresponding to the target object based on a plurality of groups of raw data;
[0036] An image allocation unit, used for using different target Histoimages as the images to be registered and the reference images;
[0037] Or, any of the target Histoimages is used as the image to be registered or the reference image.
[0038] Preferably, the target Histoimage generating unit comprises:
[0039] An initial Histoimage generating subunit, used for generating a plurality of initial Histoimages corresponding to the target object based on each set of raw data;
[0040] The correction subunit is used to perform correction operations on the initial Histoimage to generate the corresponding target Histoimage.
[0041] Preferably, the initial Histoimage generating subunit is further used to obtain annihilation events in each group of the raw data; perform a back-projection operation on the annihilation events to generate the corresponding initial Histoimage;
[0042] And / or, the correction operation includes at least one of scatter correction, attenuation correction, sensitivity correction and random correction.
[0043] Preferably, the image registration module includes:
[0044] An image segmentation unit, configured to perform image segmentation processing on the image to be registered and the reference image based on a preset segmentation algorithm to obtain a first image corresponding to the image to be registered and a second image corresponding to the reference image;
[0045] The image registration unit is used to match the first image and the second image based on a preset registration algorithm, and generate the deformation field for registering the image to be registered to the reference image.
[0046] Preferably, when only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other one is different from the PET modality.
[0047] Preferably, when only the image to be registered is the Histoimage of the PET modality, the reference image includes a CT modality image or an MRI modality image;
[0048] Or, when the only reference image is a Histoimage in the PET modality, the image to be registered includes a CT modality image or an MRI modality image.
[0049] In a third aspect, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein the processor implements the above-mentioned Histoimage-based image registration method when executing the computer program.
[0050] In a fourth aspect, a computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned Histoimage-based image registration method is implemented.
[0051] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.
[0052] The positive and progressive effects of this disclosure are:
[0053] The image registration method, system, device and medium disclosed in the present invention, in which at least one of the image to be registered and the reference image is a Histoimage of the PET modality, realizes the rapid and real-time generation of a Histoimage with good structural information during the PET scanning process. Since the Histoimage has low computational overhead and fast generation time, the Histoimage is used to replace the reconstructed image that requires a long reconstruction time in the prior art. While ensuring the accuracy of PET modality image registration, the real-time performance in the image registration of the PET modality is improved and the computational overhead is reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 A first flow chart of the image registration method based on Histoimage provided in Example 1 of the present disclosure;
[0055] Figure 2 A second flow chart of the image registration method based on Histoimage provided in Embodiment 1 of the present disclosure;
[0056] Figure 3 A third flow chart of the image registration method based on Histoimage provided in Embodiment 1 of the present disclosure;
[0057] Figure 4 A schematic diagram of a target Histoimage in the Histoimage-based image registration method provided in Example 1 of the present disclosure;
[0058] Figure 5A fourth flow chart of the image registration method based on Histoimage provided in Embodiment 1 of the present disclosure;
[0059] Figure 6 A schematic diagram of the structure of an image registration system based on Histoimage provided in Example 2 of the present disclosure;
[0060] Figure 7 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present disclosure. DETAILED DESCRIPTION
[0061] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.
[0062] Example 1
[0063] Image registration is the process of spatially matching two images of different or same modalities. Regarding registration between same modalities, it can be the registration of images between two scans of the same target object at different times, such as scanning before and after treatment, and then registering. The results can be used to analyze the treatment effect, etc.; it can also be the registration of scanned images between different target objects. For example, many technologies will involve a standard image, and then register the image of a specific target object to the standard image for easy analysis; it can also be the same target object in the same scan, in order to implement motion correction, etc., it is necessary to register images at different motion phases; for example, the target object did not move in the first five minutes of the 10-minute scan, and the head moved in the fifth minute and maintained this motion position until the end of the scan; at this time, the images of the first five minutes and the last five minutes need to be reconstructed and registered together to obtain a static image. In addition, technologies such as respiratory / heartbeat gating reconstruction also require same-modality registration.
[0064] This embodiment provides an image registration method based on Histoimage. Figure 1 As shown, the image registration method includes:
[0065] S101, obtaining an image to be registered and a reference image of a target object.
[0066] Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality.
[0067] S102, performing image segmentation processing on the image to be registered and the reference image, and performing image registration processing on the segmented image to be registered and the reference image to generate a deformation field for registering the image to be registered to the reference image, so as to achieve image registration of the image to be registered.
[0068] The target object may be an object to be scanned that requires medical imaging, such as a human body, an animal or others.
[0069] Histoimage, also known as a back-projection image, refers to a back-projection image obtained by weighted back-projecting the scan data during the PET scan along the response line using a weighting function.
[0070] PET is a non-invasive imaging technology commonly used in medical diagnosis and scientific research. It is a nuclear imaging technology (also called molecular imaging) that can show metabolic processes in the body. PET imaging technology detects gamma photon pairs emitted indirectly by positron-emitting radionuclides (also called radiopharmaceuticals, radionuclides or radiotracers). Before performing a PET scan, the tracer needs to be injected into the vein of the biologically active molecule. The sensitive detector of the PET system captures the photons emitted after the positron annihilation reaction inside the body. After image reconstruction, the tomographic distribution map of the body's positron nuclides is obtained to obtain a reconstructed image. However, this image reconstruction process is very time-consuming and computationally intensive, with poor real-time performance, and is not suitable for use in sensitive scenarios.
[0071] In the existing PET image registration methods, the directly acquired PET raw data generally needs to be processed into reconstructed images through methods such as FBP or OSEM, resulting in excessive time and computational overhead, which in turn limits the application of PET technology in time-sensitive scenarios. After obtaining the reconstructed image, image registration is required, but the time spent on registration itself is usually less than the time spent on reconstruction. Excessive reconstruction time and computational threshold reduce the practicality of PET image registration.
[0072] The present disclosure proposes a new PET image registration framework, which uses an optimized back-projection algorithm to quickly acquire PET modality images that meet the accuracy required for registration, namely, Histoimage of the PET modality, and provides a variety of registration algorithms based on iterative optimization or deep learning for rapid registration, ultimately alleviating the problem of poor real-time performance in PET image registration.
[0073] The Histoimage-based image registration method of this embodiment can quickly and in real time generate a Histoimage with good structural information during the PET scanning process. At least one of the image to be registered and the reference image is a Histoimage of the PET modality. The Histoimage is used to replace the reconstructed image that requires a long reconstruction time in the prior art. Since the Histoimage has a low computational overhead and a fast generation time, it does not require a traditional image reconstruction process, thereby reducing the computing cost. While ensuring the accuracy of PET modality image registration, the real-time performance of the PET modality image registration is improved, and the computational overhead is reduced.
[0074] In an optional embodiment, if Figure 2 As shown, the above step S101 includes:
[0075] S1011. Acquire several groups of raw data corresponding to the target object during the PET scanning process.
[0076] S1012. Generate several target Histoimages corresponding to the target objects based on several groups of raw data.
[0077] S1013, using different target Histoimages as images to be registered and reference images.
[0078] The raw data is the scanning data in the PET list mode or the scanning data in the sinogram format acquired by the scanner during the PET scanning process of the target object, specifically, the photon coincidence events used for PET modality image reconstruction.
[0079] When the raw data is in list mode, Histoimage specifically determines the physical position of one or more annihilation points based on the time difference between the two photons of the same annihilation event in the scanning data arriving at the detectors at both ends, and corresponds them to the pixels in the image domain for weighted accumulation to obtain an image.
[0080] During the PET scan, the collected raw data can be grouped according to the motion state and respiratory phase of the target object to obtain several groups of raw data corresponding to the target object, and each group of raw data can correspond to a target Histoimage.
[0081] The Histoimage-based image registration method of this embodiment can register multiple Histoimage images under the PET modality, and use different target Histoimages as the images to be registered and the reference images, that is, the images to be registered and the reference images are both Histoimages of the PET modality.
[0082] In an optional embodiment, if Figure 3 As shown, the above step S101 includes:
[0083] S1011. Acquire several groups of raw data corresponding to the target object during the PET scanning process.
[0084] S1012. Generate several target Histoimages corresponding to the target objects based on several groups of raw data.
[0085] S1014. Use any target Histoimage as an image to be registered or a reference image.
[0086] The Histoimage-based image registration method of this embodiment can also be applied to cross-modality registration between PET modality and other modalities.
[0087] For example, if an image of the PET modality is registered to an image of other modalities, and the target Histoimage is used as the image to be registered, then the reference image is the image of other modalities, and the image to be registered is the Histoimage of the PET modality; if an image of other modalities is registered to an image of the PET modality, and the target Histoimage is used as the reference image, then the image to be registered is the image of other modalities, and the reference image is the Histoimage of the PET modality; that is, during cross-modality registration, only one of the image to be registered and the reference image is the Histoimage of the PET modality.
[0088] In an optional implementation, the above step S1012 includes:
[0089] S10121. Based on each set of raw data, generate several initial Histoimages corresponding to the target object.
[0090] S10122. Perform correction operations on the initial Histoimage to generate a corresponding target Histoimage.
[0091] For each set of raw data, each response line in the raw data is back-projected to the corresponding designated area to obtain an initial Histoimage, and the initial Histoimage is corrected to obtain a target Histoimage.
[0092] According to the spatial geometric sensitivity of the PET machine and the physical properties of each photon coincidence event, the initial Histoimage is corrected using corresponding physical methods to obtain the target Histoimage.
[0093] The correction operation includes at least one of scattering correction, attenuation correction, geometric sensitivity correction and random event correction.
[0094] For example, each response line in the raw data is back-projected to a specified area to obtain a point cloud image, and the obtained point cloud image is subjected to sensitivity correction and random correction according to the geometric position of each voxel to obtain a target Histoimage.
[0095] The correction operation can improve the image quality and make the image have good structural information. By performing the correction operation on the initial Histoimage, the image quality of the target Histoimage is improved.
[0096] Figure 4The target Histoimage corresponding to the chest and abdomen scan of the target object is displayed.
[0097] The Histoimage of the PET modality used in the registration process can be an initial Histoimage obtained by directly back-projecting, or a target Histoimage after the image quality is improved by physical correction methods such as sensitivity correction and random correction after back-projection.
[0098] In an optional implementation, when the generated data is in list mode, the above step S10121 includes:
[0099] S101211. Obtain annihilation events in each set of birth data.
[0100] S101212. Perform a back-projection operation on the annihilation event to generate a corresponding initial Histoimage.
[0101] An annihilation event refers to the physical process of mass-energy conversion that occurs when a positron meets an electron, resulting in the emission of two photons (collectively called photon coincidence events) that are 180 degrees away from each other. Each annihilation event in the raw data is weighted back-projected to the corresponding designated area to generate the initial Histoimage. The specific process is to accumulate all photon coincidence events in the corresponding designated area one by one to obtain a three-dimensional graph, that is, to obtain the initial Histoimage.
[0102] The scanner used to collect raw data can be a scanner equipped with TOF (time-of-flight) function or a scanner that does not support TOF function. For scanners equipped with TOF function, the designated area can be the center of TOF bin. TOF can estimate the location of annihilation event by measuring the time difference between the two photons arriving at the two detectors. Affected by the spatial resolution of PET detector, the location of annihilation event is a range, and the center of TOF bin refers to the center of this range.
[0103] For scanners that do not support TOF function, the designated area is the center of the LOR (line of response).
[0104] In an optional embodiment, if Figure 5 As shown, the above step S102 includes:
[0105] S1021. Perform image segmentation processing on the image to be registered and the reference image based on a preset segmentation algorithm to obtain a first image corresponding to the image to be registered and a second image corresponding to the reference image.
[0106] S1022: Based on a preset registration algorithm, match the first image and the second image to generate a deformation field for registering the image to be registered to the reference image.
[0107] The first image contains corresponding segmentation information, the second image contains corresponding segmentation information, and the image segmentation processing may be organ-level image segmentation processing, such as single-organ segmentation or multi-organ segmentation.
[0108] Specifically, the segmentation information can be a mask, and multi-organ segmentation can be used to obtain masks of multiple different organs, and then use the masks to assist in registration; a first image corresponding to the image to be registered and a second image corresponding to the reference image are obtained through an image segmentation algorithm, wherein the information on the coordinate position of each organ in the first image and the second image, that is, the mask, is used to obtain masks of boundary ranges corresponding to multiple different organs, and then image registration is performed. The mask plays a guiding role in the image registration process. For example, the boundary of the organ mask is the edge of the organ, which can assist in the registration of this area and obtain a more accurate deformation field.
[0109] For example, a target area of the target object is scanned, where the target area is a certain area of the target object, for example, the target area may be certain organs or a certain body part of the target object, such as the heart or the abdominal cavity.
[0110] When the target area is the abdominal cavity, organ-level image segmentation processing is used to segment the stomach, liver, spleen and other organs in the abdominal cavity separately, and a first image including a first stomach mask, a first liver mask and a first spleen mask is obtained, and a second image including a second stomach mask, a second liver mask and a second spleen mask is obtained. Based on a preset registration algorithm, one-to-one corresponding organ masks are used to assist the registration between the first image and the second image, and finally a deformation field is generated to register the image to be registered to the reference image, thereby realizing the registration process.
[0111] The preset segmentation algorithm may be a segmentation algorithm based on deep learning or a segmentation algorithm based on image processing. The preset segmentation algorithm can use a corresponding segmentation algorithm according to different target areas of the target object to ensure accurate segmentation of all target areas of the target object.
[0112] The main processing steps based on the preset registration algorithm are: preprocessing the reference image to be processed (i.e., the second image corresponding to the reference image) and the motion image (i.e., the first image corresponding to the image to be registered), sampling a series of control points at equal intervals on the reference image, and using the positions of all control points as parameters of the parameterized model to characterize the continuous deformation field of the entire image, defining the objective function as the difference between the reference image and the registered motion image, and iteratively adjusting all parameter values of the parameterized model according to the direction of descent of the objective function according to the gradient descent method until the model converges. The objective function is composed of a similarity index, a regularization index, and DICE (dice similarity coefficient) between the organ-level masks of the two images, and the complete deformation field on the image is obtained by interpolation using the parameterized model.
[0113] The preset registration algorithm in this embodiment can be a traditional registration algorithm based on B (basis) spline, or an optimization algorithm based on other parameterized models, such as a rigid registration algorithm based on 6 degrees of freedom (rotation and displacement); it can also be a registration algorithm based on deep learning, such as a registration model based on U-net (a deep learning image segmentation network) architecture.
[0114] In an optional implementation, the two PET modality Histoimages may be homomodally registered, that is, both the image to be registered and the reference image are PET modality Histoimages.
[0115] In an optional embodiment, when only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other one is different from the PET modality.
[0116] The image registration method based on Histoimage of this embodiment can be applied to cross-modality registration between PET modality and other modalities. When only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other one is different from the PET modality, that is, other modality. It can be cross-modality registration between the Histoimage of the PET modality and other modality images such as CT / MRI.
[0117] In an optional embodiment, when the only image to be registered is a Histoimage in the PET modality, the reference image includes but is not limited to a CT modality image or an MRI modality image.
[0118] In this embodiment, the image to be registered is a Histoimage of the PET modality, and the reference image is a CT modality image or an MRI modality image. By registering the Histoimage of the PET modality to the CT modality image or the MRI modality image, image registration between different modalities is achieved.
[0119] In an optional embodiment, when the only reference image is a Histoimage in the PET modality, the images to be registered include but are not limited to CT modality images or MRI modality images.
[0120] In this embodiment, the reference image is a Histoimage of the PET modality, and the image to be registered is a CT modality image or an MRI modality image. By registering the CT modality image or the MRI modality image to the Histoimage of the PET modality, image registration between different modalities is achieved.
[0121] Example 2
[0122] This embodiment provides an image registration system based on Histoimage, and the image registration system corresponds to the image registration method based on Histoimage in Embodiment 1; Figure 6 As shown, the image registration system includes:
[0123] An image acquisition module 1 is used to acquire an image to be registered and a reference image of a target object;
[0124] Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality;
[0125] The image registration module 2 is used to perform image segmentation processing on the image to be registered and the reference image, and perform image registration processing on the segmented image to be registered and the reference image, and generate a deformation field for registering the image to be registered to the reference image to achieve image registration of the image to be registered.
[0126] In an optional embodiment, the image acquisition module 1 includes:
[0127] A raw data acquisition unit 11 is used to acquire several groups of raw data corresponding to the target object during the PET scanning process;
[0128] A target Histoimage generating unit 12 is used to generate a plurality of target Histoimages corresponding to the target object based on a plurality of sets of raw data;
[0129] An image allocation unit 13, used to use different target Histoimages as images to be registered and reference images;
[0130] Or, any target Histoimage is used as the image to be registered or the reference image.
[0131] In an optional embodiment, the target Histoimage generation unit 12 includes:
[0132] The initial Histoimage generating subunit 121 is used to generate a number of initial Histoimages corresponding to the target object based on each set of raw data;
[0133] The correction subunit 122 is used to perform correction operations on the initial Histoimage to generate a corresponding target Histoimage.
[0134] In an optional implementation, the initial Histoimage generation subunit 121 is further configured to obtain annihilation events in each set of raw data, perform back-projection operations on the annihilation events, and generate a corresponding initial Histoimage.
[0135] In an optional embodiment, the correction operation includes at least one of scatter correction, attenuation correction, sensitivity correction and random correction.
[0136] In an optional embodiment, the image registration module 2 includes:
[0137] An image segmentation unit 21 is used to perform image segmentation processing on the image to be registered and the reference image based on a preset segmentation algorithm to obtain a first image corresponding to the image to be registered and a second image corresponding to the reference image;
[0138] The image registration unit 22 is used to match the first image and the second image based on a preset registration algorithm, and generate a deformation field for registering the image to be registered to the reference image.
[0139] In an optional implementation, the image segmentation process is an organ-level image segmentation process.
[0140] In an optional embodiment, when only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other one is different from the PET modality.
[0141] In an optional embodiment, when the only image to be registered is a Histoimage of the PET modality, the reference image includes a CT modality image or an MRI modality image;
[0142] In an optional embodiment, when the only reference image is a Histoimage in the PET modality, the image to be registered includes a CT modality image or an MRI modality image.
[0143] The Histoimage-based image registration system of this embodiment has the same working principle as the Histoimage-based image registration method in Example 1, and will not be described in detail here.
[0144] The Histoimage-based image registration system of this embodiment can quickly and in real time generate a Histoimage with good structural information during a PET scan. At least one of the image to be registered and the reference image is a Histoimage of the PET modality. The Histoimage is used to replace the reconstructed image that requires a long reconstruction time in the prior art. Since Histoimage has a low computational overhead and a fast generation time, it does not require a traditional image reconstruction process, thereby reducing the computational cost. While ensuring the accuracy of PET modality image registration, the real-time performance of PET modality image registration is improved, thereby reducing the computational overhead.
[0145] Example 3
[0146] This embodiment provides an electronic device, Figure 7 The electronic device includes a memory, a processor, and a computer program stored in the memory and executed on the processor. When the processor executes the program, the image registration method based on Histoimage in the above embodiment 1 is implemented. Figure 7 The electronic device 80 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0147] like Figure 7 As shown, the electronic device 80 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 80 may include, but are not limited to: at least one processor 81, at least one memory 82, and a bus 83 connecting different system components (including the memory 82 and the processor 81).
[0148] The bus 83 includes a data bus, an address bus, and a control bus.
[0149] The memory 82 may include a volatile memory, such as a random access memory (RAM) 821 and / or a cache memory 822 , and may further include a read-only memory (ROM) 823 .
[0150] The memory 82 may also include a program / utility 825 having a set (at least one) of program modules 824, such program modules 824 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0151] The processor 81 executes computer programs stored in the memory 82 to perform various functional applications and data processing, such as the Histoimage-based image registration method in the above embodiment 1 of the present disclosure.
[0152] The electronic device 80 may also communicate with one or more external devices 84 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 85. Furthermore, the model generating device 80 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 86. Figure 7 As shown, the network adapter 86 communicates with other modules of the model-generated device 80 via the bus 83. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the model-generated device 80, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.
[0153] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.
[0154] Example 4
[0155] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the image registration method based on Histoimage in the above embodiment 1 are implemented.
[0156] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.
[0157] In a possible implementation, the present disclosure can also be implemented in the form of a program product, which includes program code. When the program product is executed on a terminal device, the program code is used to enable the terminal device to execute the steps in the Histoimage-based image registration method in Example 1 above.
[0158] Among them, the program code for executing the present disclosure can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as a separate software package, partially on the user device and partially on a remote device, or completely on the remote device.
[0159] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.
Claims
1. An image registration method based on Histoimage, characterized in that: The image registration method comprises: Acquire the image to be registered and the reference image of the target object; Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality; The image to be registered and the reference image are subjected to image segmentation processing, and the image to be registered and the reference image after the segmentation processing are subjected to image registration processing to generate a deformation field for registering the image to be registered to the reference image, so as to realize image registration of the image to be registered.
2. The image registration method according to claim 1, characterized in that: The step of acquiring the image to be registered and the reference image of the target object comprises: Acquire several groups of raw data corresponding to the target object during the PET scanning process; Generate a plurality of target Histoimages corresponding to the target object based on a plurality of groups of raw data; Using different target Histoimages as the image to be registered and the reference image; Or, any of the target Histoimages is used as the image to be registered or the reference image.
3. The image registration method according to claim 2, characterized in that: The step of generating a plurality of target Histoimages corresponding to the target object based on the plurality of sets of generated data comprises: Based on each group of raw data, generate a number of initial Histoimages corresponding to the target object; The initial Histoimage is corrected to generate the corresponding target Histoimage.
4. The image registration method according to claim 3, characterized in that: The step of generating an initial Histoimage corresponding to the target object based on each set of raw data comprises: Obtaining annihilation events in each set of the birth data; Performing a back-projection operation on the annihilation event to generate the corresponding initial Histoimage; And / or, the correction operation includes at least one of scatter correction, attenuation correction, sensitivity correction and random correction.
5. The image registration method according to claim 1, characterized in that: The step of performing image segmentation processing on the image to be registered and the reference image, performing image registration processing on the segmented image to be registered and the reference image, and generating a deformation field for registering the image to be registered to the reference image comprises: Based on a preset segmentation algorithm, performing image segmentation processing on the image to be registered and the reference image to obtain a first image corresponding to the image to be registered and a second image corresponding to the reference image; Based on a preset registration algorithm, the first image and the second image are matched to generate the deformation field for registering the image to be registered to the reference image.
6. The image registration method according to claim 1, characterized in that: When only one of the image to be registered and the reference image is a Histoimage of the PET modality, the modality of the other image is different from the PET modality.
7. The image registration method according to claim 6, characterized in that: When only the image to be registered is the Histoimage of the PET modality, the reference image includes a CT modality image or an MRI modality image; Or, when the only reference image is a Histoimage in the PET modality, the image to be registered includes a CT modality image or an MRI modality image.
8. An image registration system based on Histoimage, characterized in that: The image registration system comprises: An image acquisition module, used to acquire the image to be registered and the reference image of the target object; Wherein, at least one of the image to be registered and the reference image is a Histoimage of the PET modality; The image registration module is used to perform image segmentation processing on the image to be registered and the reference image, and perform image registration processing on the image to be registered and the reference image after the segmentation processing, and generate a deformation field for registering the image to be registered to the reference image, so as to realize image registration of the image to be registered.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, the image registration method based on Histoimage as described in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the Histoimage-based image registration method as described in any one of claims 1 to 7 is implemented.