Model training method, target element determination method, device and equipment

Abstract

The invention provides a model training method, a target element determination method, devices and equipment, and relates to the field of physics and computers, in particular to application of physics and computer technologies in the field of radiation medical treatment. The model training method comprises the steps that element density information of multiple original materials is updated based on target elements, and multiple sample materials are obtained; determining a plurality of sample CT image pairs associated with the sample materials based on the X feature parameters of the target CT equipment and the element density information of the plurality of sample materials; wherein the sample CT image pair comprises a first sample CT image containing a target element and a second sample CT image not containing the target element, and X is a positive integer; the initial model is trained according to the multiple sample CT images, a target model corresponding to the target CT equipment is obtained, and the target model is used for determining the target element content in the original CT image obtained through scanning of the target CT equipment and the target CT image after the target element is separated.

Description

Technical Field

[0001] The present disclosure relates to the fields of physics and computers, specifically to the application of physics and computer technology in the field of radiological medicine, and more specifically to a model training method, a target element determination method, a device, and an apparatus. Background Art

[0002] Computed tomography (CT) imaging technology is widely used in medical diagnosis. CT imaging and image reconstruction techniques can reveal the internal structure of an object. CT imaging can effectively improve the clarity of CT images of the substance being tested by injecting specific elements into the material. However, the resulting CT images cannot be used for subsequent calculations. Summary of the Invention

[0003] The present invention provides a model training method, a target element determination method, a device, and equipment.

[0004] According to a first aspect of the present invention, a model training method is provided, comprising: updating elemental density information of a plurality of original materials based on a target element to obtain a plurality of sample materials; determining a plurality of sample CT image pairs associated with the sample materials based on X characteristic parameters of a target CT device and the elemental density information of the plurality of sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element, where X is a positive integer; and training an initial model based on the plurality of sample CT image pairs to obtain a target model corresponding to the target CT device, wherein the target model is used to determine the target element content in the original CT image scanned by the target CT device and the target CT image after the target element is separated.

[0005] According to an embodiment of the present invention, elemental density information of multiple raw materials is updated based on a target element to obtain multiple sample materials, including: performing the following operations on each raw material: randomly adding the target element to the raw material within a preset range to obtain Y sample materials corresponding to the raw material, wherein the mass fraction of the target element in different sample materials varies, and Y is a positive integer; and determining elemental density information for each of the Y sample materials based on the mass fraction of the target element. According to an embodiment of the present invention, multiple sample CT image pairs associated with the sample materials are determined based on X characteristic parameters and elemental density information of the multiple sample materials, including: determining X*Y first sample CT images based on the X characteristic parameters and elemental density information of the Y sample materials; and removing the target element content from the elemental density information of each of the Y sample materials, and determining X*Y second sample CT images based on the elemental density information after removing the target element content.

[0006] According to an embodiment of the present invention, multiple sample CT image pairs associated with the sample materials are determined based on X characteristic parameters and elemental density information of the multiple sample materials, including: performing the following operations on each sample material: calculating a CT value of each voxel in the sample material based on the characteristic parameters and the elemental density information of the sample material; and determining a sample CT image corresponding to the sample material based on the CT value of each voxel.

[0007] According to an embodiment of the present invention, an initial model is trained based on multiple sample CT image pairs to obtain a target model corresponding to a target CT device, including: adding noise information to a first sample CT image associated with each sample material to generate enhanced sample CT image pairs associated with each sample material, wherein the number of enhanced sample CT image pairs is greater than the number of sample CT image pairs; and training the initial model based on the enhanced sample CT image pairs to obtain a target model corresponding to the target CT device.

[0008] According to an embodiment of the present invention, an initial model is trained based on multiple sample CT image pairs to obtain a target model corresponding to a target CT device, including: determining the first sample CT image in the enhanced sample CT image pair as input data of the initial model, and determining the target element content in the sample material and the second sample CT image in the enhanced sample CT image pair as output data of the initial model; and training the initial model based on the input data and the output data to obtain the target model.

[0009] According to a second aspect of the present invention, a method for determining target element content is provided, comprising: scanning a substance to be tested based on X scanning energy spectra of a target CT device, respectively, to obtain X CT images; inputting the X CT images into an elemental analysis model corresponding to the target CT device, and determining, by the elemental analysis model, the target element content in the substance to be tested and CT images that do not contain the target element based on the X CT images; wherein the elemental analysis model is a target model trained using the above-described model training method.

[0010] According to a third aspect of the present invention, a model training apparatus is provided, comprising: an acquisition module for updating elemental density information of a plurality of original materials based on a target element to obtain a plurality of sample materials; a first determination module for determining a plurality of sample CT image pairs associated with the sample materials based on X characteristic parameters of a target CT device and the elemental density information of the plurality of sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element; and a training module for training an initial model based on the plurality of sample CT image pairs to obtain a target model corresponding to the target CT device, wherein the target model is used to determine the target element content in the original CT image scanned by the target CT device and the target CT image after the target element is separated.

[0011] According to a fourth aspect of the present invention, a target element content determination apparatus is provided, comprising: a scanning module for scanning a substance to be tested based on X scanning energy spectra of a target CT device, respectively, to obtain X CT images; a second determination module for inputting the X CT images into an elemental analysis model corresponding to the target CT device, whereby the elemental analysis model determines the target element content in the substance to be tested and CT images that do not contain the target element based on the X CT images; wherein the elemental analysis model is a target model trained using the aforementioned model training apparatus.

[0012] According to a fifth aspect of the present invention, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0014] Figure 1 A flowchart of a model training method according to an embodiment of the present invention is schematically shown;

[0015] Figure 2 A flowchart schematically illustrates updating element density information of multiple original materials based on target elements to obtain multiple sample materials according to an embodiment of the present disclosure;

[0016] Figure 3 Schematically illustrates a flow chart for determining a plurality of sample CT image pairs associated with sample materials based on X characteristic parameters and element density information of a plurality of sample materials according to an embodiment of the present disclosure;

[0017] Figure 4 A flowchart schematically illustrates a method for training an initial model based on multiple sample CT images to obtain a target model corresponding to a target CT device according to an embodiment of the present disclosure;

[0018] Figure 5 is a flow chart of a method for determining target element content according to an embodiment of the present disclosure;

[0019] Figure 6 The following schematically shows a structural block diagram of a model training device according to an embodiment of the present invention;

[0020] Figure 7 A schematic block diagram of a target element content determination device according to an embodiment of the present invention is shown;

[0021] Figure 8 The block diagram of an electronic device according to an embodiment of the present invention is schematically shown. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments and the accompanying drawings. It is apparent that the embodiments described are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0023] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the presence of the features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0024] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood in a broad sense. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection, or mutual communication; direct connection or indirect connection through an intermediate medium; internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0025] In the description of the present invention, it should be understood that the terms "longitudinal", "length", "circumferential", "front", "rear", "left", "right", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the subsystem or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0026] Throughout the drawings, identical elements are represented by identical or similar reference numerals. Conventional structures or configurations may be omitted where they may obscure the understanding of the present invention. Furthermore, the shapes, sizes, and positional relationships of the components in the drawings do not reflect actual size, proportion, or actual positional relationships. Furthermore, any reference symbols placed between parentheses should not be construed as limiting.

[0027] Similarly, in order to streamline the present invention and aid in understanding one or more of the various disclosed aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. Descriptions with reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" and the like mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0028] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means at least two, such as two or three, unless otherwise specifically defined.

[0029] The embodiments of the present disclosure provide a model training method, a target element determination method, an apparatus, and a device. Before introducing the technical solutions provided by the embodiments of the present disclosure, the relevant technologies involved in the present disclosure are first described.

[0030] In some CT imaging modalities, dual-energy CT can be used to identify trace elements in the human body, such as iron, copper, and iodine, which have higher atomic numbers. Dual-energy CT refers to an energy spectrum with two or more different energy components.

[0031] By injecting specific elements into the material under test, the clarity of CT images can be improved, thereby better assisting in delineating treatment targets and organs at risk. However, the CT images obtained after injection of specific elements cannot be used for dose calculations. For one thing, the specific elements significantly increase the stopping power of the specific material, affecting the range calculation of the proton / heavy ion beam. Furthermore, specific elements are rapidly metabolized in the human body, and the time from CT image acquisition to the start of radiotherapy typically takes several days. By the time radiotherapy begins, the material under test no longer contains the specific elements. At this point, the elemental density distribution in the target area differs from that in the CT image containing the specific elements. Dose calculations based on CT images containing the specific elements can lead to errors. Therefore, it is necessary to separate the specific elements from the CT images to obtain pseudo-conventional CT images that do not contain the specific elements, and to use these images for subsequent calculations. Existing methods for separating specific elements use material decomposition, assuming that the material contains only two materials, iodine and water. Approximating the material as water is inconsistent with actual conditions and can lead to errors. This method is also susceptible to image noise and artifacts, which can lead to bias.

[0032] Figure 1 The following schematically shows a flow chart of a model training method according to an embodiment of the present invention.

[0033] like Figure 1 As shown, the model training method of this embodiment includes operations S110 to S130.

[0034] In operation S110 , element density information of a plurality of original materials is updated based on a target element to obtain a plurality of sample materials.

[0035] In some embodiments, the elemental density information of the raw material can be a priori information obtained in advance from a database. The elemental density information varies for different types of raw materials. The elemental density information for each raw material can include: the types of elements contained in the tissue material, the density of each element, and the distribution of multiple elements in the tissue material. The sample material serves as a basic database for subsequent determination of multiple sample CT image pairs.

[0036] In some embodiments, for each raw material, the mass proportion of the target element is randomly assigned within a reasonable range to obtain multiple sample materials corresponding to the raw material. The elemental density information of the sample material is different from the elemental density information of the raw material. Each raw material A can correspond to multiple sample materials B, and the number of sample materials for each raw material is the same as the number of mass proportions of the target element assigned to the raw material. For example, if the random proportion of the target element corresponding to each raw material is 100, then the number of sample materials corresponding to the raw material is also 100. Taking the raw material as bone tissue material and the target element as iodine as an example, the mass proportion of iodine is randomly assigned to the bone tissue material within a preset range to obtain 100 bone tissue materials with different mass proportions of iodine. Taking the raw material as 71 different tissue materials as an example, by randomly assigning the mass proportions of the target element to these raw materials, 7100 sample materials can be obtained.

[0037] The elemental density information of each sample material is obtained by randomly assigning a mass proportion of the target element and then updating the elemental density of the original material based on the mass proportion of the target element.

[0038] It should be noted that due to the inherent characteristics of each tissue material, the target element is not reflected in all voxels. Therefore, when constructing the sample material, it is necessary to provide a sample material with a target element mass ratio of 0% for each original material to simulate the local characteristics of the target element distribution in the real CT, thereby improving the accuracy of subsequent model training and improving the model prediction accuracy.

[0039] In operation S120 , a plurality of sample CT image pairs associated with the sample materials are determined based on the X characteristic parameters of the target CT device and element density information of the plurality of sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element.

[0040] In some embodiments, the characteristic parameters are associated with a scanning energy spectrum used by the target CT device for scanning, and different scanning energy spectra correspond to different characteristic parameters. The characteristic parameters can be used to describe the energy spectrum characteristics of the target CT device. The scanning energy spectrum of the target CT device can be used to scan a reference phantom to determine the characteristic parameters corresponding to the scanning energy spectrum, and based on the multiple characteristic parameters, sample CT image pairs associated with the sample material are determined, wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element.

[0041] In some embodiments, a tissue material is scanned using a scanning energy spectrum to obtain a CT image of the tissue material. The CT image includes a distribution of CT values for the tissue material, which can reflect the interaction between matter and photons. The CT image includes a distribution of CT values for the tissue material, which can reflect the interaction between matter and photons. The target CT device can be a CT device capable of transmitting multiple energy spectra. The target device scans the tissue material based on energy spectra with different energy components, resulting in potentially different CT images, effectively preventing tissue materials with different elemental components from having identical CT values.

[0042] For example, a dual-energy CT device is used as the target CT device. Dual-energy CT devices have two scanning energy spectra. When scanning multiple tissue samples using a single X-ray source, the elemental compositions of these samples may differ, but their CT values under a single energy spectrum may be the same. In this case, further differentiation between the two tissue samples is impossible in the CT images, leading to biased diagnostic results. Therefore, dual-energy CT can be used to compare two CT images of two tissue samples, each scanned using two different energy spectra. While two tissue samples may have the same CT values under one energy spectrum, their CT values under the other spectrum may differ, allowing for further differentiation between the two tissue samples.

[0043] Exemplarily, a virtual scanning environment of a target CT device is simulated by characteristic parameters, a virtual tissue corresponding to the sample material is simulated based on element density information of the sample material, and a sample CT image obtained by scanning the virtual tissue under the virtual scanning environment is calculated.

[0044] In some embodiments, the first and second sample CT images are calculated based on characteristic parameters of the element density distribution and the scanning energy spectrum. Based on the characteristic parameters, the scanning environment of the CT device can be simulated to determine the photon energy spectrum emitted by the target CT device. Based on the element density information, the element density at different locations in the sample material can be simulated to calculate the absorption capacity of photon energy at different locations in the sample material. Furthermore, based on the absorption capacity, the CT values at different locations in the sample material can be calculated to obtain the first and second sample CT images.

[0045] Exemplarily, multiple first sample CT images associated with the sample material are calculated based on the characteristic parameters of different scanning energy spectra and the element density information of the sample material. It should be noted that for a sample material with a target element content of 0%, its sample CT image is also marked as a first sample CT image.

[0046] and removing the target element from the element density information of each sample material, and calculating a plurality of second sample CT images associated with the sample material based on the element density information after the target element is removed.

[0047] It should be noted that the first sample CT image and the second sample CT image mentioned in the embodiments of the present disclosure are both obtained based on the characteristic parameters of at least two scanning energy spectra, that is, the first sample CT image and the second sample CT image are determined respectively using the characteristic parameters of the two scanning energy spectra. The first sample CT image refers to a plurality of CT images containing target elements obtained based on the characteristic parameters of different energy spectra, and the second sample CT image refers to a plurality of CT images not containing target elements obtained based on the characteristic parameters of different energy spectra.

[0048] The target CT device is a dual-energy CT device, and the characteristic parameter is K corresponding to the dual-energy scanning spectrum. 1,H and K 1,L For example, the first sample CT image in each sample CT image pair refers to the dual-energy first sample CT image, and the second sample CT image refers to the dual-energy second sample CT image. 1,H is the characteristic parameter of high-energy CT spectrum, K 1,L It is a characteristic parameter of the low-energy CT energy spectrum. The high-energy spectrum and low-energy spectrum here are relative concepts. The dual-energy CT device generates two sets of scanning energy spectra simultaneously. The higher energy of these two sets of scanning energy spectra is the high-energy spectrum, and the lower energy is the low-energy spectrum.

[0049] In operation S130, an initial model is trained based on multiple sample CT image pairs to obtain a target model corresponding to the target CT device, wherein the target model is used to determine the target element content in the original CT image scanned by the target CT device and the target CT image after separating the target elements.

[0050] In some embodiments, the initial model can be a fully connected neural network. The fully connected neural network is trained using multiple sample CT image pairs. A fully connected operation can be used to extract the correlation between the CT value of the first sample CT image and the target element content, as well as the mapping relationship between the first sample CT image and the second sample CT image. This allows the trained target model to determine the target element content based on the original CT image scanned by the CT device and generate a CT image (i.e., a target CT image) that does not contain the target element based on the original CT image. The original CT image refers to a CT image obtained by scanning after the target element is injected into the material under test. The target element content refers to the density-mass ratio of the target element in the material under test.

[0051] It should be noted that the energy spectrum parameters of different CT devices may be different. For each CT device, the following parameters should be used: Figure 1The model training method shown trains a target model corresponding to the CT device. The model training method provided by the embodiment of the present disclosure introduces known tissue materials as prior information, randomly increases the density and mass ratio of the target element in these tissue materials, obtains multiple sample materials, and calculates the CT images corresponding to different energy spectra of these sample materials when they contain the target element and when they do not contain the target element. The sample CT image pairs for training the model are constructed, and a machine learning method is used to train the target model using the CT images containing the target element as input data and the CT images without the target element and the mass ratio of the target element as output data. The target model is then trained to separate the target element content in the CT image through the target model, which is beneficial to improving the accuracy of subsequent operations.

[0052] Figure 2 The flowchart of obtaining multiple sample materials by respectively updating element density information of multiple original materials based on target elements according to an embodiment of the present disclosure is schematically shown.

[0053] like Figure 2 As shown, this embodiment updates element density information of multiple original materials based on target elements to obtain multiple sample materials, including operations S210 to S220, and operations S210 to S220 are performed on each original material.

[0054] In operation S210 , a target element is randomly added to an original material within a preset range to obtain Y sample materials corresponding to the original material, wherein the mass proportion of the target element in different sample materials is different, and Y is a positive integer.

[0055] In operation S220 , element density information of each of the Y sample materials is determined based on the mass fraction of the target element.

[0056] In some embodiments, the raw material and its elemental density information may be pre-acquired from a database. The raw material includes various tissue materials, such as bone tissue materials, soft tissue materials, etc., and the elemental density information of different raw materials is different.

[0057] For each tissue material, the mass proportion of the target element is randomly assigned within a preset range to obtain Y sample materials corresponding to the original material, wherein the mass proportion of the target element in different sample materials is different.

[0058] For example, if the target element is iodine and Y = 100, for each tissue material, a random iodine mass fraction is assigned within a preset range. Each tissue material can correspond to 100 random iodine mass fractions, resulting in 100 sample materials corresponding to the original material. The preset range can be, for example, 0% to 3%. It should be noted that for each tissue material, a sample material with a 0% iodine mass fraction must be included.

[0059] In some embodiments, the element density information corresponding to each sample material is updated based on the mass proportion of the target element in the sample material. For example, the mass proportion of each element in the element density information is normalized to keep the total mass proportion of each element in the sample material at 100%.

[0060] By randomly assigning target element concentrations to the raw material, we can simulate the differences in target element uptake across different tissue materials in real-world scenarios. This ensures that the training data covers the full range of conditions, from the absence of the target element to high concentrations. This effectively increases data diversity, thereby improving the generalization ability of the trained model. Furthermore, since the raw material is similar to real tissue materials, the sample materials obtained using this method can more accurately reflect the complex composition of actual tissue materials, thereby improving the accuracy of the target element separation model.

[0061] Figure 3 The flowchart of determining a plurality of sample CT image pairs associated with sample materials based on X characteristic parameters and element density information of a plurality of sample materials according to an embodiment of the present disclosure is schematically shown.

[0062] like Figure 3 As shown, this embodiment determines a plurality of sample CT image pairs associated with the sample materials based on X characteristic parameters and element density information of a plurality of sample materials, including operations S310 to S320.

[0063] In operation S310 , X*Y first sample CT images are determined based on the X feature parameters and Y element density information of the sample materials.

[0064] In operation S320 , the target element content in the element density information of each of the Y sample materials is removed, and X*Y second sample CT images are determined based on the element density information after the target element content is removed.

[0065] In some embodiments, a sample CT image of each sample material can be divided into a plurality of voxels. A voxel is the smallest unit of volume in a CT image. The CT value corresponding to each voxel represents the interaction between the material within that volume unit and photons. If the material within the voxel has a strong ability to absorb photon energy, the CT value of that voxel will also be higher.

[0066] In some embodiments, the following operations are performed for each sample material: a CT value of each voxel in the sample material is calculated based on the characteristic parameters and the elemental density information of the sample material; and a sample CT image corresponding to the sample material is determined based on the CT value of each voxel. For example, an attenuation coefficient μ of each voxel in the sample material can be calculated based on the elemental density information and the characteristic parameters of the sample material, and a CT value of each voxel is calculated based on the attenuation coefficient μ. A sample CT image of the sample material is obtained based on the CT value of each voxel in the sample material.

[0067] With the characteristic parameter k 1,H and k 2,L For example, where k 1,H is the characteristic parameter corresponding to the high scanning energy spectrum, k 2,L are the characteristic parameters corresponding to the low scan energy spectrum, based on k 1,H and k 2,L The high-energy and low-energy first sample CT images of each sample material are determined based on the element density information of each sample material, and are recorded as the dual-energy first sample CT image. It should be noted that for sample materials with a target element content of 0%, their sample CT images are also marked as the dual-energy first sample CT image.

[0068] After obtaining the dual-energy first sample CT image of each sample material, the target element content in the element density information of each sample material is removed, and based on the element density information after removing the target element content and the characteristic parameters, the high-energy and low-energy second sample CT materials of each sample material are determined and recorded as dual-energy second sample CT images, and sample CT image pairs associated with each sample material are obtained, each sample CT image pair including the dual-energy first sample CT image and the dual-energy second sample CT image of the sample material.

[0069] For example, characteristic parameters corresponding to the target CT device's scan energy spectrum can be determined using a known reference phantom. For example, a reference material is scanned using X scan energy spectra of the CT device to obtain X scan data of the reference material; and the X scan data are compared with pre-acquired reference data of the reference material to obtain X characteristic parameters of the X scan energy spectra.

[0070] The reference material may be a known standard phantom. For example, information such as the element density distribution, total material density, and CT images obtained based on a known scanning energy spectrum of the reference material are all known.

[0071] For example, the reference material can be a Gammex phantom or a CIRS phantom. The Gammex phantom is scanned using the target CT device based on M scan energy spectra, generating X CT images of the Gammex phantom as scan data. Based on the known CT images of the Gammex phantom, the elemental density distribution, and the total material density, the absorption degree of each voxel in the Gammex phantom for a specific photon energy can be determined. Based on the absorption degree and the scan data, the photon energy emitted by the CT device's emission source is calculated, thereby determining the CT device's characteristic energy spectrum.

[0072] For example, the absorption capacity of a voxel of a certain material in a Gammex phantom for radiation of a specific energy can be assumed to be fixed. Based on known CT images of the Gammex phantom, the correspondence between CT values and photon energies in the energy spectrum is determined. Based on this correspondence, the photon energies in the energy spectrum are calculated based on the CT values in the scan data. Consequently, X characteristic parameters of the scan energy spectrum can be determined based on X CT images in the scan data.

[0073] Figure 4 The flowchart of training an initial model based on multiple sample CT image pairs and obtaining a target model corresponding to a target CT device according to an embodiment of the present disclosure is schematically shown.

[0074] like Figure 4 As shown, this embodiment trains an initial model based on a plurality of sample CT image pairs to obtain a target model corresponding to a target CT device, including operations S410 to S420.

[0075] In operation S410 , noise information is added to the sample CT image pairs associated with each sample material to generate enhanced sample CT image pairs associated with each sample material, wherein the number of enhanced sample CT image pairs is greater than the number of sample CT image pairs.

[0076] In operation S420 , an initial model is trained based on the enhanced sample CT image pair to obtain a target model corresponding to the target CT device.

[0077] In some embodiments, taking into account the noise in the actual CT scanning process (such as equipment errors, environmental interference, etc.), the basic data set can be expanded based on the sample CT image pairs to obtain a model training database, wherein the model training data set includes multiple enhanced sample CT image pairs, and the enhanced sample CT image pairs are obtained by noise-adding the first sample CT image. The model trained with the model training data set can adapt to the uncertainty of the real CT scanning scene and improve the robustness of the model.

[0078] Exemplarily, assuming that the measurement error of each first sample CT image follows a Gaussian distribution, independent noise is added to the high-energy CT value and the low-energy CT value of each original first sample CT image to generate new perturbed data (i.e., the noisy first sample CT image). An enhanced sample CT image pair is constructed based on the noisy first sample CT image, completing the expansion of the sample CT image pair. Exemplarily, the second sample CT image and the target element mass fraction in the enhanced sample CT image pair remain unchanged. The enhanced sample CT image pair is obtained by establishing a corresponding relationship between the noisy first sample CT image, the second sample CT image, and the target element mass fraction.

[0079] Enhanced sample CT image pairs generated by adding noise can better simulate real-world scenarios. Model training based on these enhanced sample CT image pairs allows the model to be exposed to more noisy data during training, learning to ignore random perturbations and improving the accuracy and robustness of model predictions. Furthermore, data augmentation can expand the size of the training set and prevent model overfitting.

[0080] In some embodiments, the first sample CT image in the enhanced sample CT image pair is determined as the input data of the initial model, the target element content in the sample material and the second sample CT image in the enhanced sample CT image pair are determined as the output data of the initial model, and the initial model is trained based on the input data and the output data to obtain the target model.

[0081] In some embodiments, the first sample CT image is used as input data of the model, and the target element content and the second sample CT image are used as output data of the model, so that the model can calculate and output the target element content and the CT image without the target element based on the first sample CT image.

[0082] For example, the model can perform independent predictions based on voxels, calculating the target element density-to-mass ratio and the second CT value of each voxel using the first CT value of that voxel. The target element content of each voxel in the sample material is combined to obtain the target element content corresponding to the sample material. Furthermore, the second CT values of each voxel in the sample material are combined according to spatial position to obtain a second sample CT image of the sample material. The first CT value can be the CT value of the voxel in the first sample CT image, and the second CT value can be the CT value of the voxel in the second sample CT image.

[0083] Continuing with the example of iodine as the target element, the model uses the iodine-containing dual-energy CT value of the sample object as input data, and the iodine-free dual-energy CT value and iodine density-to-mass ratio of the sample object as output data. This allows the model to determine the iodine density-to-mass ratio and the iodine-free dual-energy CT value based on the iodine-containing dual-energy CT value of the sample object. The sample object is a voxel in the sample material.

[0084] For example, during training, the model establishes a mapping from dual-energy iodine-containing CT values to iodine density-mass ratios, and also a mapping from dual-energy iodine-containing CT values to dual-energy iodine-free CT values. The model uses dual-energy iodine-containing CT values to learn the attenuation characteristics of iodine at different energies and tissue materials. It then decomposes iodine-containing CT values into a combination of "iodine-free tissue" and "iodine," ultimately outputting iodine-free CT values and iodine density-mass ratios.

[0085] Figure 5 4 is a flow chart of a method for determining target element content according to an embodiment of the present disclosure.

[0086] like Figure 5 As shown, the target element content determination method of this embodiment includes operations S510 to S520.

[0087] In operation S510 , a substance to be tested is scanned based on X scanning energy spectra of a target CT device to obtain X CT images.

[0088] In operation S520 , the X CT images are input to an element analysis model corresponding to the target CT device, and the element analysis model determines the target element content in the substance to be tested and the CT images that do not contain the target element based on the X CT images.

[0089] In the embodiment of the present disclosure, the element analysis model is used Figures 1 to 4 The target model is obtained by training the provided model training method.

[0090] For example, X CT images containing a specific element are input into the element analysis model, and the element analysis model can output the content of the specific element and a conventional CT image that does not contain the specific element.

[0091] In the embodiment of the present disclosure, the process of determining the elemental density distribution of the substance to be tested based on M CT images through the elemental analysis model is similar to the training process described above and will not be repeated here for the sake of simplicity.

[0092] Figure 6 The structural block diagram of the model training device according to an embodiment of the present invention is schematically shown.

[0093] like Figure 6 As shown, the model training device 600 of this embodiment includes an acquisition module 610 , a first determination module 620 , and a training module 630 .

[0094] The acquisition module 610 is used to update the element density information of the plurality of original materials based on the target element to obtain a plurality of sample materials. In one embodiment, the acquisition module 610 can be used to perform the operation S110 described above, which will not be repeated here.

[0095] The first determination module 620 is configured to determine, based on the X characteristic parameters of the target CT device and the elemental density information of the plurality of sample materials, a plurality of sample CT image pairs associated with the sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element. In one embodiment, the first determination module 620 may be configured to perform operation S120 described above, which will not be further described here.

[0096] Training module 630 is configured to train an initial model based on multiple sample CT image pairs to obtain a target model corresponding to the target CT device. The target model is used to determine the distribution information content of target elements in the original CT image obtained by scanning the target CT device, as well as the target CT image after separating the target elements. In one embodiment, training module 630 can be configured to perform operation S130 described above and will not be further described here.

[0097] According to embodiments of the present invention, any multiple modules among the acquisition module 610, the first determination module 620, and the training module 630 may be combined into a single module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. According to embodiments of the present invention, at least one of the acquisition module 610, the first determination module 620, and the training module 630 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware through any other reasonable means of circuit integration or packaging, or may be implemented in any one of software, hardware, and firmware, or any suitable combination of these. Alternatively, at least one of the acquisition module 610, the first determination module 620, and the training module 630 may be at least partially implemented as a computer program module that, when executed, performs the corresponding functionality.

[0098] Figure 7 The structure block diagram of the target element content determination device according to an embodiment of the present invention is schematically shown.

[0099] like Figure 7 As shown, the target element content determination device 700 of this embodiment includes a scanning module 710 and a second determination module 720 .

[0100] The scanning module 710 is used to scan the material to be tested based on the X scanning energy spectra of the target CT device to obtain X CT images. In one embodiment, the scanning module 710 can be used to perform the operation S510 described above, which will not be repeated here.

[0101] Second determination module 720 is configured to input the X CT images into an elemental analysis model corresponding to the target CT device. The elemental analysis model then determines the target element content in the test material and the CT images that do not contain the target element based on the X CT images. In one embodiment, second determination module 720 may be configured to perform operation S520 described above, and will not be further described here.

[0102] Among them, the element analysis model is used Figure 6 The target model obtained by training the device shown.

[0103] According to embodiments of the present invention, any multiple modules in the scanning module 710 and the second determination module 720 may be combined into a single module, or any one of them may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. According to embodiments of the present invention, at least one of the scanning module 710 and the second determination module 720 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware through any other suitable means of circuit integration or packaging, or may be implemented in any one of software, hardware, and firmware, or any suitable combination thereof. Alternatively, at least one of the scanning module 710 and the second determination module 720 may be at least partially implemented as a computer program module that, when executed, performs the corresponding functionality.

[0104] Figure 8 The block diagram of an electronic device according to an embodiment of the present invention is schematically shown.

[0105] like Figure 8As shown, an electronic device 800 according to an embodiment of the present invention includes a processor 801, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 8602 or programs loaded from a storage unit 808 into a random access memory (RAM) 803. The processor 801 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or related chipsets and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 801 may also include onboard memory for caching purposes. The processor 801 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0106] The RAM 803 stores various programs and data required for the operation of the electronic device 800. The processor 801, ROM 802, and RAM 803 are connected to each other via a bus 804. The processor 801 executes the programs in the ROM 802 and / or RAM 803 to perform the various operations of the method flow according to the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than the ROM 802 and RAM 803. The processor 801 may also execute the programs stored in one or more memories to perform the various operations of the method flow according to the embodiment of the present invention.

[0107] According to an embodiment of the present invention, electronic device 800 may further include an input / output (I / O) interface 805, which is also connected to bus 804. Electronic device 800 may also include one or more of the following components connected to I / O interface 805: an input section 806 including a keyboard, mouse, etc.; an output section 807 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 808 including a hard disk; and a communication section 809 including a network interface card such as a LAN card or modem. Communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to I / O interface 805 as needed. Removable media 811, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 810 as needed, so that computer programs read from the removable media can be installed into storage section 808 as needed.

[0108] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not be incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, and when executed, implements the method according to the embodiments of the present disclosure.

[0109] According to embodiments of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium. Examples include, but are not limited to, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0110] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Those skilled in the art may make various substitutions and modifications without departing from the scope of the present invention, and such substitutions and modifications are intended to be included within the scope of protection of the present invention.

Claims

1. A model training method, characterized in that: The method comprises: updating element density information of a plurality of original materials based on target elements to obtain a plurality of sample materials; determining a plurality of sample CT image pairs associated with the sample materials based on X characteristic parameters of the target CT device and element density information of the plurality of sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element, and X is a positive integer; An initial model is trained based on the multiple sample CT image pairs to obtain a target model corresponding to the target CT device, wherein the target model is used to determine the target element content in the original CT image scanned by the target CT device and the target CT image after separating the target element.

2. The model training method according to claim 1, characterized in that The step of updating element density information of a plurality of original materials based on the target element to obtain a plurality of sample materials includes: For each source material, do the following: Randomly adding a target element to the original material within a preset range to obtain Y sample materials corresponding to the original material, wherein the mass proportion of the target element in different sample materials is different, and Y is a positive integer; The element density information of each of the Y sample materials is determined based on the mass proportion of the target element.

3. The model training method according to claim 2, characterized in that The determining of a plurality of sample CT image pairs associated with the sample materials based on the X characteristic parameters and the element density information of the plurality of sample materials comprises: Determine X*Y first sample CT images based on the X characteristic parameters and the element density information of the Y sample materials respectively; The target element content in the element density information of each of the Y sample materials is removed, and X*Y second sample CT images are determined based on the element density information after the target element content is removed.

4. The model training method according to claim 3, characterized in that The determining of a plurality of sample CT image pairs associated with the sample materials based on the X characteristic parameters and the element density information of the plurality of sample materials comprises: For each sample material, do the following: Calculating a CT value of each voxel in the sample material based on the characteristic parameters and element density information of the sample material; A sample CT image corresponding to the sample material is determined based on the CT value of each voxel.

5. The model training method according to claim 1, characterized in that The training of the initial model based on the plurality of sample CT images to obtain a target model corresponding to the target CT device includes: adding noise information to the first sample CT image associated with each sample material respectively, to generate enhanced sample CT image pairs associated with each sample material, wherein the number of the enhanced sample CT image pairs is greater than the number of the sample CT image pairs; An initial model is trained based on the enhanced sample CT image pair to obtain a target model corresponding to the target CT device.

6. The model training method according to claim 5, characterized in that The training of the initial model based on the plurality of sample CT images to obtain a target model corresponding to the target CT device includes: Determining the first sample CT image in the enhanced sample CT image pair as input data of the initial model, and determining the target element content in the sample material and the second sample CT image in the enhanced sample CT image pair as output data of the initial model; The initial model is trained based on the input data and the output data to obtain a target model.

7. A method for determining target element content, characterized in that: include: Scanning the material to be tested based on X scanning energy spectra of the target CT device to obtain X CT images; Inputting the X CT images into an element analysis model corresponding to the target CT device, and determining, by the element analysis model, the target element content in the substance to be tested and the CT images that do not contain the target element based on the X CT images; Wherein, the element analysis model is a target model trained using the method according to any one of claims 1-6.

8. A model training device comprising: An acquisition module, configured to update element density information of a plurality of original materials based on a target element to obtain a plurality of sample materials; a first determining module configured to determine a plurality of sample CT image pairs associated with the sample materials based on X characteristic parameters of the target CT device and element density information of the plurality of sample materials; wherein the sample CT image pairs include a first sample CT image containing the target element and a second sample CT image not containing the target element; A training module is used to train an initial model based on the multiple sample CT image pairs to obtain a target model corresponding to the target CT device, wherein the target model is used to determine the target element content in the original CT image scanned by the target CT device and the target CT image after separating the target element.

9. A device for determining target element content, comprising: A scanning module, configured to scan the material to be tested based on X scanning energy spectra of the target CT device to obtain X CT images; a second determination module, configured to input the X CT images into an element analysis model corresponding to the target CT device, and determine, by the element analysis model, the target element content in the substance to be tested and the CT images that do not contain the target element based on the X CT images; Wherein, the element analysis model is a target model trained using the device as described in claim 8.

10. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6 and / or claim 7.

Images