An electromagnetic sparse imaging method for metal foreign body in human body fusing background knowledge

Abstract

The present application belongs to the medical imaging technical field, especially to a kind of electromagnetic sparse imaging method of metal foreign matter in human body fusing background knowledge, the present application includes the following steps: construct the electromagnetic scattering model of human tissue without metal foreign matter: construct the electromagnetic scattering model of human tissue after embedding metal foreign matter: fuse background knowledge, establish difference scattering model to extract metal foreign matter response: define contrast source and construct inversion model based on joint sparse regularization: solve inversion model and reconstruct metal foreign matter image;The present application proposes a modeling method fusing background prior knowledge, the proposed method can utilize the prior knowledge of human tissue, which helps to improve the problem of ill-conditioned, in order to further overcome inverse ill-conditioned, use joint sparse regularization technique to constrain the solving process of electromagnetic inverse problem, further improve the positioning accuracy of metal foreign matter, improve imaging accuracy, propose a kind of joint sparse regularization method, which can enhance the accuracy of imaging.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, specifically to an electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge. Background Technology

[0002] In recent years, with the frequent occurrence of incidents such as industrial production accidents, traffic accidents, accidental injuries, and clinical surgeries, cases of injuries caused by embedded or residual metallic foreign bodies in the human body have been gradually increasing. These metallic foreign bodies may originate from metal fragments, explosion debris, surgical instrument residue, or damaged implants, and are often characterized by small size, irregular shape, deep embedding, and uncertain number. These characteristics greatly increase the difficulty for clinicians to achieve accurate localization and rapid intervention within a limited timeframe, especially in emergency treatment and resource-constrained environments, creating an urgent need for efficient and reliable detection technologies.

[0003] Currently, commonly used clinical methods for detecting metallic foreign bodies mainly include X-rays, computed tomography (CT), and ultrasound imaging. While these methods are effective in routine medical conditions, they have significant limitations in on-site rescue, ambulance transport, and other emergency medical scenarios: X-ray and CT equipment are bulky, complex to operate, and both emit ionizing radiation. Ultrasound image quality depends heavily on the operator's skill level, making it difficult for non-professionals to obtain stable and reliable results. Therefore, the adaptability and practicality of existing imaging technologies in emergency and mobile medical situations are significantly limited. In contrast, electromagnetic imaging technology exhibits unique advantages in detecting metallic foreign bodies in the human body. It is highly sensitive to metallic targets, has good tissue penetration capabilities, and poses no risk of ionizing radiation. With the gradual miniaturization and integration of hardware systems, electromagnetic imaging equipment possesses greater portability and on-site deployment capabilities, providing a feasible and efficient technical approach for rapid localization and assisted diagnosis and treatment. However, existing electromagnetic imaging methods still face many challenges when applied to the detection of metallic foreign bodies within the human body. First, the complex structure and non-uniform spatial distribution of human tissues significantly alter the propagation path and scattering characteristics of electromagnetic waves, increasing uncertainties in modeling and imaging. Second, electromagnetic imaging suffers from pathological issues, making the inversion process extremely unstable, leading to large reconstruction errors and even failure to converge. Therefore, effectively integrating background knowledge of human tissues during electromagnetic imaging and improving the stability and accuracy of inversion through sparse imaging techniques to achieve high-resolution localization and detection of metallic foreign bodies within the human body has become one of the core technical challenges urgently needing to be addressed in this field.

[0004] Therefore, we propose an electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge to address the above problems. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an electromagnetic sparse imaging method for metallic foreign objects within the human body that incorporates background knowledge. This method considers prior knowledge of the human tissue background, which helps improve imaging accuracy. To enhance the stability of the inversion process, a joint sparse regularization technique is used to constrain the solution process of the electromagnetic inverse problem, further improving the localization accuracy of metallic foreign objects and solving the problems mentioned in the background technology.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention specifically adopts the following technical solution:

[0009] An electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge includes the following steps:

[0010] S1. Construct an electromagnetic scattering model of human tissue without metallic foreign objects:

[0011] Establish the first state equation and the first data equation to describe the electromagnetic scattering behavior of human tissue.

[0012] S2. Construct an electromagnetic scattering model of human tissue after embedding a metallic foreign object:

[0013] Establish a second state equation and a second data equation to describe the electromagnetic scattering behavior when a metallic foreign object is present;

[0014] S3. Integrate background knowledge to establish a differential scattering model to extract the response of metallic foreign objects:

[0015] S4. Define the comparison source and construct an inversion model based on joint sparse regularization:

[0016] S5. Solve the inversion model and reconstruct the image of the metallic foreign object.

[0017] Further, in step S1, the first state equation is: The first data equation is: ;

[0018] in, The total electric field within the imaging region, The incident field generated by the transmitting antenna. The scattered field is measured by the receiving antenna. For a mapping operator defined within the imaging region, its diagonal elements are: Its off-diagonal elements are ; To define the mapping operator from the imaging region to the receiving antenna, The contrast of human tissues is known prior knowledge.

[0019] Furthermore, the contrast of the human tissue Determined based on the known relative permittivity and conductivity of various human tissues at the imaging frequency.

[0020] Further, in step S2, the second state equation is: ;

[0021] The second data equation is: ;

[0022] in, = + , This indicates the contrast change caused by the introduction of a metallic foreign object.

[0023] Further, in step S3, the second state equation is subtracted from the first state equation to obtain the third state equation:

[0024]

[0025] in, , ;

[0026] Subtracting the first data equation from the second data equation yields the third data equation:

[0027]

[0028] in: .

[0029] Further, in step S4, the comparison source is defined: ;

[0030] For an antenna array containing N shared transmit and receive antennas, all measured scattering data are organized into a matrix. Organize the contrast sources illuminated by all antennas into a matrix W; construct the following loss function for solution:

[0031]

[0032] in, Representing a matrix -norm, For regularization parameters, Here is a multi-task sparsity metric function, defined as: ;or ;

[0033] Where M is the total number of grids after discretizing the imaging region, and N is the number of antennas.

[0034] Further, in step S4, the multi-task sparsity measurement function This is used to promote the group sparsity of the contrast source on the spatial grid, so that the reconstruction energy is concentrated in the boundary region of the metallic foreign object.

[0035] Furthermore, the antenna array contains at least four antennas, N, arranged circumferentially around the part of the human body to be imaged.

[0036] Furthermore, in step S5, the loss function is solved using a relaxation iteration method. The source distribution matrix was obtained. ;

[0037] For the comparison source distribution matrix The amplitudes of all antenna illuminations are summed to obtain the spatial distribution of the metal foreign object boundary, thereby enabling its localization and imaging.

[0038] Furthermore, in step S5, the relaxation iteration method is one of the following: iterative shrinkage threshold algorithm, alternating direction multiplier method, or proximal gradient method.

[0039] (III) Beneficial Effects

[0040] Compared with existing technologies, this invention provides an electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge, and has the following beneficial effects:

[0041] This invention proposes a modeling method that integrates prior knowledge of the background. This method can utilize prior knowledge of human tissue, which helps to improve the ill-conditioning of the problem. To further overcome inverse ill-conditioning, a joint sparse regularization technique is used to constrain the solution process of the electromagnetic inverse problem, further improving the positioning accuracy of metallic foreign objects and enhancing imaging accuracy. A joint sparse regularization method is proposed, which can enhance imaging accuracy. Attached Figure Description

[0042] Figure 1 This is a flowchart of the present invention.

[0043] Figure 2 This is a schematic diagram of the imaging scene of the present invention;

[0044] Figure 3 This is a schematic diagram of the imaging results of a metallic foreign object embedded in the edge of the human body using the method proposed in this invention.

[0045] Figure 4 This is a schematic diagram of the imaging results of a metallic foreign object embedded in the human body using the method proposed in this invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Example 1:

[0048] like Figure 1-4 As shown, an embodiment of the present invention proposes an electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge, comprising the following steps:

[0049] S1. Based on prior knowledge of human tissues, a state equation and data equation are proposed to describe their electromagnetic scattering behavior.

[0050] First, we model the electromagnetic scattering problem of human tissue that does not contain metallic foreign objects. The state equation and data equation describing its electromagnetic scattering behavior are as follows:

[0051] (1)

[0052] (2)

[0053] in, The total electric field within the imaging region, The incident field generated by the transmitting antenna. The scattered field is measured by the receiving antenna. For a mapping operator defined within the imaging region, its diagonal elements are: Its off-diagonal elements are ; To define the mapping operator from the imaging region to the receiving antenna, The contrast of human tissues is known prior knowledge.

[0054] When a metallic foreign object is embedded in the human body, its scattering behavior is modeled again. The state equation and data equation describing its electromagnetic scattering behavior are as follows:

[0055] (3)

[0056] (4)

[0057] According to electromagnetic theory, a metallic foreign object can be considered as a high-loss dielectric scatterer. Therefore, the contrast of a metallic foreign object embedded in the human body can be expressed as... .

[0058] In order to utilize prior knowledge of the human body background, subtracting equation (1) from equation (3) yields:

[0059] (5)

[0060] in: = -1 , .

[0061] Furthermore, by subtracting equation (2) from equation (4), we can obtain:

[0062] (6)

[0063] in: , = .

[0064] Because there is a nonlinear relationship between the scattered field data and the contrast to be reconstructed, a nonlinear inversion algorithm is required for solving the problem. Nonlinear algorithms typically have high computational complexity and require significant computational resources, which cannot meet real-time requirements and limits their engineering application in rapid detection of metallic foreign objects.

[0065] S2. An electromagnetic sparse imaging method based on background fusion is proposed to improve the accuracy of locating metallic foreign objects.

[0066] According to the comparison source As can be seen from the definition and the boundary conditions satisfied by the metallic target, compared to the source only Non-zero at the boundary of the metallic target.

[0067] Therefore, by reconstructing the intensity distribution of the contrast source, the geometric contour of the metallic target can be further inferred. Although equation (6) is linear, its solution suffers from severe ill-conditioning, and direct solution often leads to unstable or non-unique results. Therefore, it is necessary to introduce a suitable regularization method to improve the stability and accuracy of the reconstruction. Since the contrast source is only distributed in the boundary region of the metallic target, the spatial distribution has significant sparsity. Therefore, a sparse regularization strategy can be introduced to improve the stability and resolution of the reconstruction results.

[0068] In electromagnetic imaging, antenna arrays are typically arranged around the human body, with each antenna element sharing both transmitting and receiving capabilities. Here, we assume the number of antenna elements is... All measured scattering data are stored in a matrix: T represents the transpose of the matrix, and the corresponding contrast source for each antenna illumination is stored in the matrix: Considering illumination from all antennas, the loss function can be defined as follows:

[0069] (7)

[0070] in, Representing a matrix -norm, For regularization parameters, These are sparsity control parameters. The defined multi-task sparsity metric is as follows:

[0071] (8)

[0072] Where N is the total number of discrete grids in the imaging region.

[0073] Equation (7) can be solved using the relaxation iteration method, and then the determination can be made. Finally, the amplitudes of the contrast sources illuminated by all antennas are summed to restore the boundary of the metallic target.

[0074] This invention proposes a modeling method that integrates prior knowledge of the background. This method can utilize prior knowledge of human tissue, which helps to improve the ill-conditioning of the problem. To further overcome inverse ill-conditioning, a joint sparse regularization technique is used to constrain the solution process of the electromagnetic inverse problem, further improving the positioning accuracy of metallic foreign objects and enhancing imaging accuracy. A joint sparse regularization method is proposed, which can enhance imaging accuracy.

[0075] Example 2:

[0076] To verify the accuracy and practicality of the localization algorithm proposed in Example 1, simulation tests were conducted on two typical scenarios with distributed metallic foreign objects. During the tests, the imaging area was 0.1m × 0.1m in size, and the imaging frequency was 350MHz. Four transmitting and receiving antennas were uniformly arranged on a circle with a radius of 0.12m, and the transmitting antenna used a uniform plane wave. For numerical solution, the imaging area was uniformly discretized into a 40 × 40 square grid. At a frequency of 350MHz, the relative permittivity of skin was 48.1, and its conductivity was 0.666 S / m; the relative permittivity of muscle was 57.6, and its conductivity was 0.783 S / m; and the relative permittivity of bone was 12, and its conductivity was 0.178 S / m. The scattered field data was obtained through simulation using the method of moments (MoM), and 10% Gaussian white noise was added to the generated scattered field data for electromagnetic imaging. The metallic foreign object was circular with a radius of 5mm.

[0077] like Figure 3 As shown, in the first test scenario, the metallic foreign object was located at the edge of the human tissue, with coordinates (2.5mm, 0mm), so electromagnetic waves could easily penetrate the human tissue.

[0078] The location of the reconstructed metallic foreign object is as follows: Figure 2 As shown, the proposed electromagnetic sparse imaging method that integrates background prior knowledge can accurately reconstruct the location of metallic foreign objects; the computation time for reconstructing metallic foreign objects is only 0.341s, which indicates that the proposed method has very high computational efficiency and can meet the requirements of real-time imaging.

[0079] Furthermore, Figure 4 A more complex test was conducted where the size of the metallic foreign object remained constant, and the object was located inside human tissue with its center at coordinates (0 mm, 0 mm). The deep embedding of the metallic foreign object reduced the penetration of electromagnetic waves; however, the reconstruction results showed that the proposed method could accurately locate the metallic foreign object with a computation time of 0.335 s.

[0080] In specific embodiments, the choice of imaging frequency is flexible; as long as the selected frequency ensures that electromagnetic waves can penetrate human tissue, the proposed method is applicable. Furthermore, the number of antennas can be flexibly set. Finally, in the above specific embodiment, a forearm model is selected as the human tissue; in fact, the proposed method can be used for other human torsos or other biological tissues.

[0081] In summary, this invention proposes a modeling method that integrates prior knowledge of the background. This method can utilize prior knowledge of human tissues, which helps to improve the ill-conditioning of the problem. To further overcome inverse ill-conditioning, a joint sparse regularization technique is used to constrain the solution process of the electromagnetic inverse problem, further improving the positioning accuracy of metallic foreign objects and enhancing imaging accuracy. A joint sparse regularization method is proposed, which can enhance imaging accuracy.

[0082] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An electromagnetic sparse imaging method for metallic foreign bodies in the human body that incorporates background knowledge, characterized in that: Includes the following steps: S1. Construct an electromagnetic scattering model of human tissue without metallic foreign objects: Establish the first state equation and the first data equation to describe the electromagnetic scattering behavior of human tissue. S2. Construct an electromagnetic scattering model of human tissue after embedding a metallic foreign object: Establish a second state equation and a second data equation to describe the electromagnetic scattering behavior when a metallic foreign object is present; S3. Integrate background knowledge to establish a differential scattering model to extract the response of metallic foreign objects: S4. Define the comparison source and construct an inversion model based on joint sparse regularization: S5. Solve the inversion model and reconstruct the image of the metallic foreign object.

2. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S1, the first state equation is: The first data equation is: ; in, The total electric field within the imaging region, The incident field generated by the transmitting antenna. The scattered field is measured by the receiving antenna. For a mapping operator defined within the imaging region, its diagonal elements are: Its off-diagonal elements are ; To define the mapping operator from the imaging region to the receiving antenna, The contrast of human tissues is known prior knowledge.

3. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 2, characterized in that: The contrast of the human tissue Determined based on the known relative permittivity and conductivity of various human tissues at the imaging frequency.

4. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S2, the second state equation is: ; The second data equation is: ; in, = + , This indicates the contrast change caused by the introduction of a metallic foreign object.

5. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S3, the second state equation is subtracted from the first state equation to obtain the third state equation: in, , ; Subtracting the first data equation from the second data equation yields the third data equation: in: .

6. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S4, the comparison source is defined: ; For an antenna array containing N shared transmit and receive antennas, all measured scattering data are organized into a matrix. Organize the contrast sources illuminated by all antennas into a matrix W; construct the following loss function for solution: in, Representing a matrix -norm, For regularization parameters, Here is a multi-task sparsity metric function, defined as: ;or Where M is the total number of grids after discretizing the imaging region, and N is the number of antennas.

7. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 6, characterized in that: In step S4, the multi-task sparsity measurement function This is used to promote the group sparsity of the contrast source on the spatial grid, so that the reconstruction energy is concentrated in the boundary region of the metallic foreign object.

8. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 6, characterized in that: The antenna array has at least four antennas, N, arranged circumferentially around the part of the human body to be imaged.

9. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S5, the loss function is solved using a relaxation iteration method. The source distribution matrix was obtained. ; For the comparison source distribution matrix The amplitudes of all antenna illuminations are summed to obtain the spatial distribution of the metal foreign object boundary, thereby enabling its localization and imaging.

10. The electromagnetic sparse imaging method for metallic foreign bodies in the human body that integrates background knowledge according to claim 1, characterized in that: In step S5, the relaxation iteration method is one of the following: iterative shrinkage threshold algorithm, alternating direction multiplier method, or proximal gradient method.

Images