Method and device for judging whether to apply DIBH technology or not based on intelligently generated image

Abstract

The invention discloses a method and device for judging whether to apply a DIBH technology based on an intelligent generated image, and the method comprises the steps: carrying out the feature vector extraction and model training of a deformation field between free breathing and deep inspiration breath-holding state pairing images, and generating a group principal component analysis model; the method comprises the following steps: by comparing a positioning image of a patient in an individualized deep inspiration and breath-holding state, optimizing principal component feature vector weights of a model and a deformation field, enabling the deformation field to act on a computed tomography image of the patient in a free breathing state, and generating individualized sDIBH-CT; on the basis of the sDIBH-CT, the average heart dose is calculated through radiotherapy target area sketching and plan design, and the reduction value of the average heart dose relative to the free breathing state is obtained; and comparing the reduced value with a preset benefit index to finish screening whether the patient is suitable for the DIBH technology or not. According to the method and the system, the possible benefits of the patient applying the DIBH are obtained in advance, and effective auxiliary decisions are ensured to be obtained clinically.

Description

Technical Field

[0001] The present disclosure relates to deep inspiration breath-hold application technology, and in particular to a method, apparatus, device, and medium for determining whether to apply the DIBH technology based on intelligently generated images. Background Art

[0002] Breast radiotherapy after breast-conserving surgery can reduce breast cancer recurrence and mortality. However, patients receiving radiotherapy for left-sided breast cancer may be at increased risk for heart and coronary artery disease. For every 1 Gy increase in the average cardiac dose during radiotherapy, the coronary artery risk increases by 7.4%, with no apparent threshold. Clinically, deep inspiration breath-hold (DIBH) has been shown to expand the lungs and move the heart away from the chest wall, effectively reducing the dose to the heart and lungs and minimizing target positioning uncertainty caused by respiratory motion. It is an important radiotherapy technique for protecting the heart and lungs. However, its application requires not only hardware support but also patient compliance and tolerance. It also requires additional training and treatment time, which is a significant burden in routine clinical practice. Furthermore, individual differences in body mass index, lung function, respiratory control ability, and anatomical features are also important factors in determining whether a patient is suitable for DIBH.

[0003] Current clinical practice generally uses repeated CT scans with free breathing and deep inspiration breath-hold to determine the actual benefit of DIBH. However, this method exposes patients to repeated radiation exposure. Alternatively, anatomical features such as the heart contact distance or maximum heart depth can be used to estimate cardiac radiation dose during treatment and predict the benefit of DIBH. However, the predictive accuracy of these methods needs to be improved.

[0004] Therefore, one or more methods are needed to solve the above problems.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the Invention

[0006] The purpose of the present disclosure is to provide a method, apparatus, device and medium for determining whether to apply DIBH technology based on intelligently generated images, thereby overcoming one or more problems caused by limitations and defects of related technologies to at least a certain extent.

[0007] According to one aspect of the present disclosure, a method for determining whether to apply the DIBH technology based on intelligently generated images is provided, comprising:

[0008] Based on principal component analysis, a group principal component analysis model was generated by extracting feature vectors and training models of the deformation field between paired images of free breathing and deep inspiration and breath-hold states.

[0009] Based on the population principal component analysis model, by comparing the individualized deep inspiration breath-hold positioning images of breast radiotherapy patients, the weight parameters of the principal component eigenvectors of the deformation field are optimized. The deformation field is then applied to the computed tomography images of the patients in the free-breathing state to generate individualized deep inspiration breath-hold computed tomography images.

[0010] Based on the individualized deep inspiration breath-hold computed tomography images, the average cardiac dose is calculated through radiotherapy target volume delineation and plan design to obtain a reduction value of the average cardiac dose relative to a free breathing state;

[0011] Based on the reduction value of the average cardiac dose relative to the free breathing state, screening index information is generated by comparing with a preset benefit index. Based on the screening index information, breast radiotherapy patients are screened for whether they are suitable for deep inspiration breath-hold technology treatment.

[0012] In an exemplary embodiment of the present disclosure, feature vector extraction and model training are performed on the deformation field between paired images of free breathing and deep inspiration and breath-hold state, including:

[0013] Based on the group training set, free-breathing CT images and deep inspiration breath-hold CT images were paired to generate paired deformation fields.

[0014] Based on the paired deformation field, the principal component feature vector is generated by extracting the features when the free breathing state is switched to the deep inspiration and breath holding state;

[0015] Based on the linear combination of the principal component eigenvectors, the deformation field solution formula is generated by establishing the solution formula of the paired deformation field

[0016] in, is the mean vector, u k is the characteristic vector of the dynamic motion pattern of the lungs when switching from the free breathing state to the deep inspiration and breath holding state, w k is the characteristic parameter of principal component analysis, K is a constant;

[0017] Based on the deformation field solution formula, a group principal component analysis model is generated by analyzing the respiratory motion in the group training set.

[0018] In an exemplary embodiment of the present disclosure, by comparing the scouting images of breast radiotherapy patients in the deep inspiration breath-hold state, the weight parameters of the principal component eigenvectors of the deformation field are optimized, including:

[0019] Based on the group principal component analysis model, a global loss function is generated by comparing the deep inspiration breath-hold projection images and deep inspiration breath-hold positioning images of a group of breast radiotherapy patients and calculating the feature matching loss function of the deformation field;

[0020] Based on the population principal component analysis model, a control volume loss function is generated by introducing a control volume when calculating the feature matching loss function;

[0021] By integrating the global loss function and the control volume loss function, a characteristic parameter loss function is generated.

[0022] In an exemplary embodiment of the present disclosure, by comparing the scouting images of breast radiotherapy patients in the deep inspiration breath-hold state, the weight parameters of the principal component eigenvectors of the deformation field are optimized, including:

[0023] Based on the gradient descent algorithm, the step size of iterative optimization is established through the Armijo rule to generate the iterative step size;

[0024] Based on the characteristic parameter loss function, by iteratively optimizing the weight parameters of the principal component characteristic vectors, when the characteristic parameter loss function reaches a preset decrease value, generating iterative characteristic parameters;

[0025] Based on the characteristic parameter loss function, iteratively optimizing the weight parameters of the principal component characteristic vectors, and generating iterative characteristic parameters when the number of iterations is consistent with the iteration step size;

[0026] Based on the iterative characteristic parameters, the paired deformation field is calculated using the population principal component analysis model, and the deformation field is applied to a computed tomography image of the patient in a free-breathing state to generate an individualized deep inspiration breath-hold computed tomography image.

[0027] In an exemplary embodiment of the present disclosure, the average cardiac dose is calculated by radiotherapy target volume delineation and plan design, including:

[0028] Based on the free-breathing computed tomography image, calculating the average cardiac dose by delineating the radiotherapy target volume and designing the plan in the free-breathing state to generate a free-breathing average cardiac dose;

[0029] Based on the individualized deep inspiration breath-hold computed tomography image, calculating the average cardiac dose by delineating the radiotherapy target volume and planning the individualized deep inspiration breath-hold state, thereby generating an individualized deep inspiration breath-hold average cardiac dose;

[0030] By comparing the free-breathing average heart dose and the individualized deep inspiration breath-hold average heart dose, a reduction value of the average heart dose relative to the free-breathing state is generated.

[0031] In an exemplary embodiment of the present disclosure, the preset benefit indicators include:

[0032] generating a first indicator when the free-breathing average heart dose is less than 4 Gy and the individualized deep inspiration breath-hold average heart dose is reduced by not less than 50% compared with the free-breathing average heart dose;

[0033] generating a second indicator when the free-breathing average heart dose is less than 4 Gy and the reduction of the individualized deep inspiration breath-hold average heart dose compared to the free-breathing average heart dose is less than 50%;

[0034] generating a third indicator when the free-breathing average heart dose is not less than 4 Gy and the individualized deep inspiration breath-hold average heart dose is reduced by not less than 30% compared with the free-breathing average heart dose;

[0035] When the free-breathing average heart dose is not less than 4 Gy, and the reduction of the individualized deep inspiration breath-hold average heart dose compared to the free-breathing average heart dose is less than 30%, a fourth indicator is generated.

[0036] In an exemplary embodiment of the present disclosure, comparing with a preset benefit indicator includes:

[0037] When the average cardiac dose reduction value meets the first indicator, the breast radiotherapy patient is treated with a deep inspiration breath-hold technique;

[0038] When the mean cardiac dose reduction value meets the second indicator, the breast radiotherapy patient is treated with a free breathing technique;

[0039] When the mean cardiac dose reduction value meets the third indicator, the breast radiotherapy patient is treated with a deep inspiration breath-hold technique;

[0040] When the mean cardiac dose reduction value meets the fourth indicator, the breast radiotherapy patient is treated with the free breathing technique.

[0041] In one aspect of the present disclosure, a device for determining whether to apply the DIBH technique based on intelligently generated images is provided, comprising:

[0042] The group principal component analysis model building module generates a group principal component analysis model by extracting feature vectors and training models of the deformation field between paired images of free breathing and deep inspiration and breath-hold state based on the principal component analysis method;

[0043] A weight parameter optimization module is used to optimize the weight parameters of the principal component eigenvectors of the deformation field by comparing the positioning images of breast radiotherapy patients in the deep inspiration breath-hold state. The deformation field is then applied to the CT images of the patients in the free breathing state to generate individualized deep inspiration breath-hold CT images.

[0044] The average cardiac dose calculation module is used to calculate the average cardiac dose through radiotherapy target area delineation and plan design, and obtain the reduction value of the average cardiac dose relative to the free breathing state;

[0045] The treatment technology screening module is used to generate screening indicator information by comparing with preset benefit indicators, and based on the screening indicator information, complete the screening of whether breast radiotherapy patients are suitable for deep inspiration breath-hold technology treatment.

[0046] In one aspect of the present disclosure, there is provided an electronic device, comprising:

[0047] processor; and

[0048] A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the method according to any one of the above items.

[0049] In one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to any one of the above items is implemented.

[0050] Based on the embodiment of the present disclosure, first, based on the principal component analysis method, the deformation field between the paired images of the free breathing and deep inspiration breath-hold state is extracted with feature vectors and model training to generate a group principal component analysis model. Afterwards, based on the group principal component analysis model, by comparing the positioning image of the patient's individualized deep inspiration breath-hold state, the weight parameters of the principal component feature vectors of the deformation field are optimized, and the deformation field is applied to the computed tomography image of the patient in the free breathing state to generate an individualized sDIBH-CT. Then, based on the sDIBH-CT, the average cardiac dose is calculated by delineating the radiotherapy target area and planning design to obtain the reduction value of the average cardiac dose relative to the free breathing state. Finally, the reduction value is compared with the preset benefit index to complete the screening of whether the patient is suitable for DIBH technology treatment. Therefore, the embodiments of the present disclosure provide a screening scheme for breast radiotherapy patients suitable for DIBH treatment, which can know in advance the possible benefits of DIBH application for patients without performing DIBH simulation positioning, and achieve the accuracy of screening patients for DIBH treatment, so as to ensure that effective auxiliary decision-making can be obtained during DIBH clinical treatment, thereby better leveraging the advantages of DIBH and saving medical resources.

[0051] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0052] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0054] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0055] Figure 1 This is a flow chart of a method for determining whether to apply the DIBH technology based on intelligently generated images according to one embodiment of the disclosed method;

[0056] Figure 2 A decision logic flow chart of a method for determining whether to apply the DIBH technology based on intelligently generated images according to an embodiment of the disclosed method;

[0057] Figure 3 This is a structural block diagram of a device for determining whether to apply the DIBH technology based on intelligently generated images according to an embodiment of the disclosed method;

[0058] Figure 4 A block diagram of an electronic device according to an embodiment of the disclosed method. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. Like reference numerals in the drawings represent like or similar parts, and thus repetitive description thereof will be omitted.

[0060] In addition, the described features, structures or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, materials, devices, steps, etc. can be adopted. In other cases, well-known structures, methods, devices, implementations, materials or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0061] The blocks shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. Specifically, these functional entities may be implemented in software, or in one or more software-hardened modules, or in different networks and / or processor devices and / or microcontroller devices.

[0062] First, based on the principal component analysis method, the deformation field between the paired images of free breathing and deep inspiration breath-hold state is extracted with feature vectors and model training to generate a group principal component analysis model. Afterwards, based on the group principal component analysis model, by comparing the positioning images of the patient in the individualized deep inspiration breath-hold state, the principal component feature vector weight parameters of the deformation field are optimized to generate an individualized sDIBH-CT. Then, based on the sDIBH-CT, the average cardiac dose is calculated by radiotherapy target area delineation and plan design to obtain the reduction value of the average cardiac dose relative to the free breathing state. Finally, the reduction value is compared with the preset benefit index to complete the screening of whether the patient is suitable for DIBH technology treatment. The present invention realizes the early acquisition of the possible benefits of DIBH for patients through the above method, so as to ensure that effective auxiliary decision-making can be obtained in the clinical practice of DIBH, thereby better leveraging the advantages of DIBH and saving medical resources.

[0063] In the embodiment of the present disclosure, a method for determining whether to apply DIBH technology based on intelligently generated images is first provided; in particular, a method for screening breast radiotherapy patients for application of DIBH technology based on individually generated computed tomography images is provided. Figure 1As shown in , the method for determining whether to apply the DIBH technology based on intelligently generated images may include the following steps:

[0064] Step S110, generating a group principal component analysis model by extracting feature vectors and performing model training on the deformation field between paired images of free breathing and deep inspiration and breath-hold states based on principal component analysis;

[0065] Step S120, based on the population principal component analysis model, by comparing the individualized deep inspiration breath-hold scouting images of breast radiotherapy patients, optimizing the weight parameters of the principal component eigenvectors of the deformation field, applying the deformation field to the CT images of the patients in the free-breathing state, and generating individualized deep inspiration breath-hold CT images;

[0066] Step S130, calculating the average cardiac dose based on the individualized deep inspiration breath-hold computed tomography image through radiotherapy target volume delineation and plan design to obtain a reduction value of the average cardiac dose relative to a free breathing state;

[0067] Step S140 , based on the reduction value of the average cardiac dose relative to the free breathing state, by comparing with a preset benefit index, screening index information is generated, and based on the screening index information, the breast radiotherapy patient is screened for whether the deep inspiration breath-hold technique is suitable for treatment.

[0068] The term "intelligently generated image" in this application refers to an individualized deep inspiration breath-hold computed tomography (sDIBH-CT) image generated from free-breathing images, using a population principal component analysis model and a positioning image of the individual patient during deep inspiration breath-hold. It is not a simple image, but rather a fusion of population data and individual characteristics, capable of predicting a patient's deep inspiration breath-hold image and used to determine the applicability of DIBH technology.

[0069] Below, as Figure 2 As shown, a method for determining whether to apply the DIBH technology based on intelligently generated images in an embodiment of the present disclosure will be further described.

[0070] In step S110, based on the principal component analysis method, a group principal component analysis model may be generated by extracting feature vectors and performing model training on the deformation field between paired images of free breathing and deep inspiration and breath-hold states.

[0071] In some optional embodiments of this example, the principal component analysis model is able to apply inherent regularization to its representation while providing the best linear representation of the data in terms of least squares.

[0072] Furthermore, compared to other AI-based methods, the principal component analysis model can generate deep inspiration breath-hold computed tomography images more simply and with more rational principles. Furthermore, in the present invention, the deep inspiration breath-hold state is relatively stable and repeatable, avoiding the limitations of the principal component analysis model in predicting irregular motion (such as irregular changes in respiratory rate and amplitude during continuous breathing).

[0073] Based on this, this embodiment establishes a principal component analysis model suitable for individualized patients by selecting a group training set. The specific steps are as follows:

[0074] First, the free-breathing computed tomography (FB-CT) images in the training dataset are paired with the rigidly registered deep inspiration breath-hold CT images (DIBH-CT) to find the deformation field of the pairings when the respiratory state changes. The features that change from the free-breathing state to the deep inspiration breath-hold state during pairing are extracted to generate the principal component feature vector.

[0075] Then, a set of characteristic parameters of principal component analysis (PCA) is used to characterize the paired deformation field, so that it can be approximately represented as a linear combination of the mean vector and a series of eigenvectors. The solution formula for the paired deformation field is thus established:

[0076] in, is the mean vector, u k is the characteristic vector of the dynamic motion pattern of the lungs when switching from the free breathing state (FB) to the deep inspiration and breath holding state, w k The principal component analysis characteristic parameters are used, and K is a constant. The formula is used to characterize the image volume changes and lung dynamic motion from free breathing state to deep inspiration and breath holding state.

[0077] Finally, based on the deformation field solution formula, the changes in image volume and dynamic motion of the lungs when the breathing state of individuals in the group training set changes can be analyzed to generate a group principal component analysis model.

[0078] In step S120, based on the group principal component analysis model, the weight parameters of the principal component eigenvectors of the deformation field can be optimized by comparing the positioning images of the individualized deep inspiration breath-hold state of breast radiotherapy patients. The deformation field is then applied to the computed tomography images of the patients in the free breathing state to generate individualized deep inspiration breath-hold computed tomography images.

[0079] In some optional embodiments of this example, the principal component analysis (PCA) feature parameters in the above-mentioned group principal component analysis model are iteratively optimized using the patient's individualized deep inspiration breath-hold positioning image.

[0080] Specifically, the deep inspiration breath-hold scouting image of each patient in the test set was first scanned. A linear relationship was assumed between the generated deep inspiration breath-hold projection image and the deep inspiration breath-hold scouting image. The deep inspiration breath-hold projection images of a group of breast radiotherapy patients were compared with the deep inspiration breath-hold scouting images, and the feature matching loss function for the deformation field was calculated.

[0081] Therefore, the calculation formula of the feature matching loss function between the two is established:

[0082]

[0083]

[0084] where U is a principal component analysis (PCA) eigenvector matrix, w is a vector containing the PCA eigenparameters to be optimized, φ is the parameterized paired deformation field, f0 is the scanned free-breathing CT projection image, f is the reconstructed deep-inspiration breath-hold CT image, p is the scout image in the deep-inspiration breath-hold state, and R is the orthogonal projection operator used to calculate f.

[0085] According to the calculation formula, while ensuring that the formula satisfies Under the condition of , the scanned free breathing projection image is reconstructed into a deep inspiration breath-hold computed tomography image through the parameterized paired deformation field, and matched with the positioning image under the deep inspiration breath-hold state. Then the feature matching loss function between the two is calculated through the linear relationship, that is, the weighted global loss function J is calculated. G (w).

[0086] Afterwards, to further ensure the accuracy and effectiveness of the optimized principal component analysis (PCA) feature parameters in describing individual respiratory motion characteristics, feature matching was performed by introducing pre-identified control volumes (CVs). The loss function calculation formula for the feature parameters with control volumes was established:

[0087] min·J(w)=J G (w)+λJ CV (w)

[0088]

[0089] Among them, J(w) is the feature parameter loss function, J G (w) is the weighted global loss function, J CV (w) is the control volume loss function, and λ is a constant parameter used to determine the relative weights of the weighted global loss function and the control volume loss function. For the control volume loss function, M CVis a matrix that represents the position of the control volume in the orthogonal projection image. Similarly, the above feature parameter loss function calculation formula also needs to satisfy conditions.

[0090] In some optional embodiments of this example, according to the above-mentioned feature parameter loss function, the principal component analysis feature parameters are optimized. First, the optimization algorithm formula is established:

[0091]

[0092] That is, using the gradient descent algorithm with variable w, w, φ and f are gradually converged. The iterative step size μ n Determined by Armijo's rules.

[0093] In the formula, the iteration step μ n Multiply by the partial derivative of the characteristic parameter loss function with respect to the characteristic parameter of the principal component analysis before optimization, and then multiply by the characteristic parameter w before optimization n By making the difference, the optimized principal component analysis characteristic parameter w can be calculated n+1 .

[0094] Then, based on the chain rule, through the intermediate variables The feature parameter loss function J that has no direct connection with the principal component analysis feature parameter w before optimization n Contact us. It is a linear combination of the spatial gradients of the reference image weighted by the deformation field, and can be evaluated at eight adjacent grid points to determine the next optimization direction.

[0095] However, since the principal component analysis feature parameters and the corresponding deformation field are updated in this step, it is necessary to obtain the reconstructed image f by trilinear interpolation. n+1 Accordingly, It must also be consistent with the interpolation process to obtain the correct gradient.

[0096] Finally, based on the above formula algorithm, when the characteristic parameter loss function reaches a preset drop value (with zero as the lower limit), or the number of iterations is consistent with the iteration step size, the optimization process is stopped. And based on the currently optimized principal component analysis characteristic parameters (i.e., iterative characteristic parameters) and the individual patient's free-breathing CT image, the individualized paired deformation field is calculated through the above-mentioned group principal component analysis model, and the deformation field is applied to the patient's free-breathing CT image to generate an individualized deep inspiration breath-hold CT image. In this way, the patient's individualized respiratory characteristics are introduced into the group principal component analysis model, solving the influence of the patient's individualized lung function and respiratory amplitude on DIBH-CT generation, and improving the accuracy of the prediction.

[0097] In step S130, the average cardiac dose may be calculated based on the individualized deep inspiration breath-hold computed tomography image through radiotherapy target area delineation and plan design to obtain a reduction value of the average cardiac dose relative to the free breathing state.

[0098] In some optional embodiments of this example, each patient in the group training set underwent CT simulation positioning of both FB and sDIBH states, and the radiotherapy target area was delineated and a treatment plan was designed. The treatment plan adopts the concept of hybrid intensity-modulated radiotherapy (H-IMRT) and is completed through an automatic planning script. After the treatment plan is completed, independent senior physicists and physicians calculate and review the average cardiac dose in the free breathing state and the average cardiac dose in the individualized deep inspiration breath-hold state.

[0099] The calculated free-breathing mean cardiac dose and the individualized deep inspiration breath-hold mean cardiac dose were then compared to determine the patient's benefit from switching from one respiratory state to another (i.e., the reduction in mean cardiac dose relative to the free-breathing state).

[0100] In step S140, based on the reduction value of the average cardiac dose relative to the free breathing state, screening index information can be generated by comparing with a preset benefit index. Based on the screening index information, whether breast radiotherapy patients are suitable for deep inspiration breath-hold technology treatment is completed.

[0101] In some optional embodiments of this example, first, a preset benefit indicator is established, wherein:

[0102] The first indicator is that the average heart dose during free breathing is less than 4 Gy, and the average heart dose during individualized deep inspiration breath-hold is reduced by at least 50% compared with the average heart dose during free breathing;

[0103] The second indicator is that the average heart dose during free breathing is less than 4 Gy, and the average heart dose during individualized deep inspiration breath hold is reduced by less than 50% compared with the average heart dose during free breathing;

[0104] The third indicator is that the average heart dose during free breathing is not less than 4 Gy, and the average heart dose during individualized deep inspiration breath-hold is reduced by not less than 30% compared with the average heart dose during free breathing;

[0105] The fourth indicator is that the average heart dose during free breathing is not less than 4 Gy, and the average heart dose during individualized deep inspiration breath hold is reduced by less than 30% compared with the average heart dose during free breathing.

[0106] The patient's benefit from switching respiratory states is then assessed. Specifically, if the average cardiac dose reduction after switching a patient's respiratory state meets one of the first and third indicators, the patient is considered suitable for DIBH. If the average cardiac dose reduction after switching a patient's respiratory state meets one of the second and fourth indicators, FB is considered. This segmented approach allows for more rational benefit assessment.

[0107] It should be noted that although the steps of the method disclosed herein are depicted in a particular order in the accompanying drawings, this does not require or imply that the steps must be performed in that particular order, or that all steps must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one, and / or one step may be decomposed into multiple steps.

[0108] In addition, in this exemplary embodiment, a device for determining whether to apply the DIBH technology based on intelligently generated images is also provided. Figure 3 As shown, the apparatus 300 for determining whether to apply DIBH technology based on intelligently generated images may include: a population principal component analysis model establishment module 310, a weight parameter optimization module 320, an average cardiac dose calculation module 330, and a treatment technology screening module 340. Among them:

[0109] The group principal component analysis model building module 310 generates a group principal component analysis model by extracting feature vectors and performing model training on the deformation field between paired images of free breathing and deep inspiration and breath-hold states based on principal component analysis.

[0110] The weight parameter optimization module 320 is configured to optimize the weight parameters of the principal component eigenvectors of the deformation field by comparing the individualized deep inspiration breath-hold scouting images of breast radiotherapy patients, and then apply the deformation field to the CT images of the patients in the free-breathing state to generate individualized deep inspiration breath-hold CT images.

[0111] The average heart dose calculation module 330 is used to calculate the average heart dose by delineating the radiotherapy target area and planning the radiotherapy, and obtain a reduction value of the average heart dose relative to the free breathing state;

[0112] The treatment technology screening module 340 is used to generate screening index information by comparing with preset benefit indicators, and complete the screening of whether breast radiotherapy patients are suitable for deep inspiration breath-hold technology treatment based on the screening index information.

[0113] The apparatus for determining whether to apply the DIBH technique based on intelligently generated images according to the embodiments of the present disclosure corresponds to the method for determining whether to apply the DIBH technique based on intelligently generated images according to the embodiments of the present disclosure, and the relevant contents can be referenced to each other and will not be repeated here. The beneficial technical effects corresponding to the apparatus for determining whether to apply the DIBH technique based on intelligently generated images according to the embodiments of the present disclosure can be referred to the corresponding beneficial technical effects in the corresponding exemplary method section above and will not be repeated here.

[0114] It should be noted that although the above detailed description mentions several modules or units of the apparatus 300 for determining whether to apply the DIBH technique based on intelligently generated images, this division is not mandatory. In fact, according to embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in a single module or unit. Conversely, the features and functions of a single module or unit described above can be further divided and embodied by multiple modules or units.

[0115] Below, reference Figure 4 The electronic device according to the embodiment of the present disclosure is described. The electronic device may be either or both of the first device and the second device, or a standalone device independent of them, and the standalone device may communicate with the first device and the second device to receive collected input signals from them.

[0116] Figure 4 A block diagram of an electronic device according to an embodiment of the present disclosure is illustrated.

[0117] like Figure 4 As shown, the electronic device includes one or more processors and memory.

[0118] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0119] The memory may store one or more computer program products, and the memory may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program products may be stored on the computer-readable storage medium, and the processor may execute the computer program products to implement the various embodiments and methods of the present disclosure described above and / or other desired functions.

[0120] In one example, the electronic device may further include an input device and an output device, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0121] In addition, the input device may also include, for example, a keyboard, a mouse, and the like.

[0122] The output device can output various information to the outside, including determined distance information, direction information, etc. The output device can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0123] Of course, to simplify, Figure 4 Only some of the components related to the present disclosure in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0124] In addition to the above-mentioned methods and devices, an embodiment of the present disclosure may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present disclosure described in the above part of this specification.

[0125] The computer program product may be written in any combination of one or more programming languages to implement the operations of the disclosed embodiments, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0126] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, causes the processor to execute the steps of the method according to various embodiments of the present disclosure described in the above part of this specification.

[0127] The computer-readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0128] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and not restrictive, and should not be construed as necessarily possessed by each embodiment of the present disclosure. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present disclosure to necessarily being implemented using these specific details.

[0129] Each embodiment in this specification is described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. References to the same or similar parts between the various embodiments are sufficient. For system embodiments, since they largely correspond to method embodiments, their description is relatively simple. For relevant parts, references to the description of the method embodiments are sufficient.

[0130] The block diagrams of the devices, devices, equipment, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0131] The methods and apparatus of the present disclosure may be implemented in many ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers recording media that store programs for executing the methods according to the present disclosure.

[0132] It should also be noted that in the apparatus, device, and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0133] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0134] The above description has been provided for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for determining whether to apply DIBH technology based on intelligently generated images, characterized in that: include: Based on principal component analysis, a group principal component analysis model was generated by extracting feature vectors and training models of the deformation field between paired images of free breathing and deep inspiration and breath-hold states. Based on the population principal component analysis model, by comparing the individualized deep inspiration breath-hold positioning images of breast radiotherapy patients, the weight parameters of the principal component eigenvectors of the deformation field are optimized. The deformation field is then applied to the computed tomography images of the patients in the free-breathing state to generate individualized deep inspiration breath-hold computed tomography images. Based on the individualized deep inspiration breath-hold computed tomography images, the average cardiac dose is calculated through radiotherapy target volume delineation and plan design to obtain a reduction value of the average cardiac dose relative to a free breathing state; Based on the reduction value of the average cardiac dose relative to the free breathing state, screening index information is generated by comparing with a preset benefit index. Based on the screening index information, breast radiotherapy patients are screened for whether they are suitable for deep inspiration breath-hold technology treatment.

2. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 1, characterized in that: The feature vector extraction and model training are performed on the deformation field between paired images of free breathing and deep inspiration and breath-hold state, including: Based on the group training set, free-breathing CT images and deep inspiration breath-hold CT images were paired to generate paired deformation fields. Based on the paired deformation field, the principal component feature vector is generated by extracting the features when the free breathing state is switched to the deep inspiration and breath holding state; Based on the linear combination of the principal component eigenvectors, the deformation field solution formula is generated by establishing the solution formula of the paired deformation field in, is the mean vector, u k is the characteristic vector of the dynamic motion pattern of the lungs when switching from the free breathing state to the deep inspiration and breath holding state, w k is the characteristic parameter of principal component analysis, K is a constant; Based on the deformation field solution formula, a group principal component analysis model is generated by analyzing the respiratory motion in the group training set.

3. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 1, characterized in that: By comparing the positioning images of breast radiotherapy patients in the deep inspiration breath-hold state, the weight parameters of the principal component eigenvectors of the deformation field are optimized, including: Based on the group principal component analysis model, a global loss function is generated by comparing the deep inspiration breath-hold projection images and deep inspiration breath-hold positioning images of a group of breast radiotherapy patients and calculating the feature matching loss function of the deformation field; Based on the population principal component analysis model, a control volume loss function is generated by introducing a control volume when calculating the feature matching loss function; By integrating the global loss function and the control volume loss function, a characteristic parameter loss function is generated.

4. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 3, characterized in that: By comparing the positioning images of breast radiotherapy patients in the deep inspiration breath-hold state, the weight parameters of the principal component eigenvectors of the deformation field are optimized, including: Based on the gradient descent algorithm, the step size of iterative optimization is established through the Armijo rule to generate the iterative step size; Based on the characteristic parameter loss function, by iteratively optimizing the weight parameters of the principal component characteristic vectors, when the characteristic parameter loss function reaches a preset decrease value, generating iterative characteristic parameters; Based on the characteristic parameter loss function, iteratively optimizing the weight parameters of the principal component characteristic vectors, and generating iterative characteristic parameters when the number of iterations is consistent with the iteration step size; Based on the iterative characteristic parameters, the paired deformation field is calculated using the group principal component analysis model, and the deformation field is applied to the patient's free-breathing computed tomography image to generate an individualized deep inspiration breath-hold computed tomography image.

5. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 1, characterized in that: Calculation of mean cardiac dose through radiotherapy target volume delineation and plan design, including: Based on the free-breathing computed tomography image, calculating the average cardiac dose by delineating the radiotherapy target volume and designing the plan in the free-breathing state to generate a free-breathing average cardiac dose; Based on the individualized deep inspiration breath-hold computed tomography image, calculating the average cardiac dose by delineating the radiotherapy target volume and planning the individualized deep inspiration breath-hold state, thereby generating an individualized deep inspiration breath-hold average cardiac dose; By comparing the free-breathing average heart dose and the individualized deep inspiration breath-hold average heart dose, a reduction value of the average heart dose relative to the free-breathing state is generated.

6. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 1, characterized in that: The preset benefit indicators include: generating a first indicator when the free-breathing average heart dose is less than 4 Gy and the individualized deep inspiration breath-hold average heart dose is reduced by not less than 50% compared with the free-breathing average heart dose; generating a second indicator when the free-breathing average heart dose is less than 4 Gy and the reduction of the individualized deep inspiration breath-hold average heart dose compared to the free-breathing average heart dose is less than 50%; generating a third indicator when the free-breathing average heart dose is not less than 4 Gy and the individualized deep inspiration breath-hold average heart dose is reduced by not less than 30% compared with the free-breathing average heart dose; When the free-breathing average heart dose is not less than 4 Gy, and the reduction of the individualized deep inspiration breath-hold average heart dose compared to the free-breathing average heart dose is less than 30%, a fourth indicator is generated.

7. The method for determining whether to apply the DIBH technique based on intelligently generated images according to claim 6, characterized in that: Comparisons with pre-specified benefit indicators include: When the average cardiac dose reduction value meets the first indicator, the breast radiotherapy patient is treated with a deep inspiration breath-hold technique; When the mean cardiac dose reduction value meets the second indicator, the breast radiotherapy patient is treated with a free breathing technique; When the mean cardiac dose reduction value meets the third indicator, the breast radiotherapy patient is treated with a deep inspiration breath-hold technique; When the mean cardiac dose reduction value meets the fourth indicator, the breast radiotherapy patient is treated with the free breathing technique.

8. A device for determining whether to apply DIBH technology based on intelligently generated images, characterized in that: include: The group principal component analysis model building module generates a group principal component analysis model by extracting feature vectors and training models of the deformation field between paired images of free breathing and deep inspiration and breath-hold state based on the principal component analysis method; A weight parameter optimization module is used to optimize the weight parameters of the principal component eigenvectors of the deformation field by comparing the positioning images of breast radiotherapy patients in the deep inspiration breath-hold state. The deformation field is then applied to the CT images of the patients in the free breathing state to generate individualized deep inspiration breath-hold CT images. The average cardiac dose calculation module is used to calculate the average cardiac dose through radiotherapy target area delineation and plan design, and obtain the reduction value of the average cardiac dose relative to the free breathing state; The treatment technology screening module is used to generate screening indicator information by comparing with preset benefit indicators, and based on the screening indicator information, complete the screening of whether breast radiotherapy patients are suitable for deep inspiration breath-hold technology treatment.

9. An electronic device, characterized in that: include: a memory for storing a computer program product; A processor is configured to execute the computer program product stored in the memory, and when the computer program product is executed, implements the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method described in any one of claims 1 to 7 is implemented.

Images