A deep tissue injury detection method and system based on dielectric spectrum analysis

Abstract

The present application relates to the technical field of biomedical detection, in particular to a deep tissue damage detection method and system based on dielectric spectrum analysis. The method comprises the following steps: obtaining the complex permittivity spectrum of the pelvic tissue to be detected; establishing an improved bipolar Cole-Cole model; establishing a mapping relationship between the tissue dielectric parameters and the damage degree; establishing a damage judgment function; obtaining the judgment result of the tissue damage grade and damage type; and realizing the damage detection of deep tissue. The present application uses the time-frequency joint sparse reconstruction algorithm and the deep generative adversarial network to realize the high-precision inversion of the deep tissue complex permittivity spectrum; through the construction of the improved bipolar Cole-Cole model, the layered dielectric characteristics of the tissue are accurately characterized; through the establishment of the damage judgment function, combined with the interactive analysis of the damage characteristic function, the accurate classification of single damage and composite damage is realized. The method effectively improves the sensitivity, specificity and clinical applicability of deep tissue damage detection.

Description

Technical Field

[0001] This invention relates to the field of biomedical detection technology, specifically to a method and system for detecting deep tissue damage based on dielectric spectrum analysis. Background Technology

[0002] Early detection of deep tissue injuries is a major challenge in clinical medicine, especially in the pelvic region where the deep location and complex structure make it difficult for traditional imaging methods to achieve highly sensitive and specific diagnoses. Tissue damage leads to disruption of cell membrane integrity, changes in ion concentration, and abnormal water distribution, which in turn cause alterations in dielectric properties such as dielectric constant and conductivity. Dielectric spectrum analysis technology, by detecting the electromagnetic response characteristics of tissues over a wide frequency range, provides a new approach for the non-invasive detection of deep tissue injuries.

[0003] Existing technologies mainly employ single-frequency impedance measurement or the classic Cole-Cole model for tissue dielectric analysis, which has the following drawbacks: First, it is difficult to effectively separate the signal aliasing between deep and superficial tissues, resulting in large errors in dielectric parameter extraction; second, the existing Cole-Cole model has insufficient fitting accuracy over a wide frequency range, making it difficult to characterize the non-Debye relaxation characteristics of deep tissues as they change with depth; third, there is a lack of multi-layer mapping models between dielectric parameters and damage degree, resulting in low accuracy in determining damage level and type; and fourth, the ability to analyze the interaction effects of complex damage is insufficient, failing to meet the needs of precise clinical diagnosis.

[0004] Currently, there is not enough research on the detection of deep tissue damage, and there is no specific method for detecting deep tissue damage based on dielectric spectrum analysis. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for detecting deep tissue damage based on dielectric spectrum analysis.

[0006] In a first aspect, the present invention provides a method for detecting deep tissue damage based on dielectric spectrum analysis, comprising the following steps: obtaining the complex dielectric constant spectrum of the pelvic tissue to be tested; establishing an improved bipolar Cole-Cole model based on the complex dielectric constant spectrum; establishing a mapping relationship between tissue dielectric parameters and damage degree using the improved bipolar Cole-Cole model; establishing a damage determination function for the pelvic tissue to be tested based on the mapping relationship; obtaining the damage level and damage type determination results of the pelvic tissue to be tested using the damage determination function; and realizing deep tissue damage detection through the determination results. This invention achieves high-precision quantitative characterization of the dielectric properties of deep tissues by acquiring the complex dielectric constant spectrum of the pelvic tissue under test, providing a reliable data foundation for improving the sensitivity and specificity of damage detection. By establishing an improved bipolar Cole-Cole model, it overcomes the shortcomings of existing models in terms of insufficient fitting accuracy over a wide frequency range, providing a more accurate mathematical description of the dielectric behavior of pelvic tissues. By establishing a mapping relationship between tissue dielectric parameters and damage degree, it reveals the correlation between changes in dielectric properties and the pathological state of the tissue, creating conditions for quantitative assessment of damage. By constructing a damage judgment function, it realizes intelligent damage classification through multi-parameter fusion, providing an objective basis for distinguishing different damage levels and types. Through the determination results of damage level and damage type, it achieves non-invasive and rapid detection of deep tissue damage, providing a novel technical means for the diagnosis and treatment monitoring of clinical pelvic diseases.

[0007] Optionally, obtaining the complex permittivity spectrum of the pelvic tissue to be tested includes: transmitting a swept-frequency electromagnetic wave signal to the pelvic tissue to be tested to obtain a transmitted signal and a reflected signal; designing a time-frequency joint sparse reconstruction algorithm based on the transmitted signal and the reflected signal; using the time-frequency joint sparse reconstruction algorithm to obtain the signal separation results of the pelvic tissue to be tested and the surface tissue; establishing a deep generative adversarial network dielectric inversion model based on the signal separation results; and obtaining the complex permittivity spectrum of the pelvic tissue to be tested at different depths through the deep generative adversarial network dielectric inversion model. This invention achieves efficient separation of deep tissue signals from surface tissue signals by transmitting swept-frequency electromagnetic waves to the pelvic tissue under test and acquiring transmitted and reflected signals. Combined with a time-frequency joint sparse reconstruction algorithm, it lays the foundation for improving the accuracy of deep tissue dielectric parameter extraction. By designing a time-frequency joint sparse reconstruction algorithm for signal separation, it effectively suppresses interference caused by the aliasing of signals from multiple tissues, providing important technical support for accurate dielectric analysis of pelvic tissue structures. By establishing a deep generative adversarial network dielectric inversion model, it achieves high-precision inversion of the spectrum of complex dielectric constants at different depths, providing a basis for tomographic detection of tissue damage.

[0008] Optionally, establishing the improved bipolar Cole-Cole model based on the complex permittivity spectrum includes: obtaining depth-dependent relaxation time distribution characteristics based on the complex permittivity spectrum; and establishing the improved bipolar Cole-Cole model based on the relaxation time distribution characteristics. This invention extracts depth-dependent relaxation time distribution characteristics from the complex permittivity spectrum, providing important parameters for more accurately characterizing the complex dielectric relaxation behavior of deep tissues; by analyzing the correlation between relaxation time distribution characteristics and tissue depth, it reveals the differences in dielectric response of microstructures at different tissue depths, laying a theoretical foundation for establishing a dielectric model with depth resolution; by constructing the improved bipolar Cole-Cole model, it achieves accurate fitting of the dielectric spectrum of multilayer tissues, providing an important method for analyzing the multi-scale dielectric properties of deep tissues; by incorporating depth-dependent relaxation characteristics into the bipolar Cole-Cole model, it significantly improves the model's sensitivity to pathological changes in deep tissues, providing an important technical approach for the detection and differential diagnosis of early minor injuries.

[0009] Optionally, the improved bipolar Cole-Cole model satisfies the following expression:

[0010] ,

[0011] in, For different depths and different frequencies The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This is the static ionic conductivity. The dielectric constant is the vacuum dielectric constant. This invention, by introducing a depth-dependent DC dielectric constant, can accurately characterize the differences in static polarization characteristics of tissues at different depths, providing important parameter basis for the analysis of the layered dielectric properties of deep tissues. By setting the relaxation time as a depth-related variable, it effectively reflects the dynamic relaxation behavior of tissue microstructure with depth changes, creating conditions for revealing the dielectric response mechanism of pathological changes in deep tissues. By setting the distribution coefficient, it significantly improves the model's fitting ability to the non-Debye relaxation characteristics of deep tissues, providing a mathematical basis for the accurate analysis of the complex dielectric constant spectrum. By integrating static ionic conductivity and the high-frequency limiting dielectric constant, a complete depth-resolved dielectric model system is constructed, laying a theoretical model support for the development of quantitative detection technology for deep tissue damage with clinical application value.

[0012] Optionally, establishing the mapping relationship between tissue dielectric parameters and damage degree using the improved bipolar Cole-Cole model includes: establishing a mapping model between tissue dielectric parameters and damage degree using the improved bipolar Cole-Cole model; and obtaining the mapping relationship between the tissue dielectric parameters and the damage degree through the mapping model. This invention achieves precise quantification of tissue damage degree by establishing a mapping model between deep tissue dielectric parameters and damage degree, exhibiting higher accuracy, adaptability, and real-time performance. By obtaining the dynamic mapping relationship between tissue dielectric parameters and damage degree, it realizes an intelligent conversion from dielectric measurement to clinical assessment, providing an important technical means for real-time monitoring and prognostic assessment of deep tissue damage.

[0013] Optionally, the mapping model between the tissue dielectric parameters and the degree of damage satisfies the following expression:

[0014] ,

[0015] in, For depth The degree of tissue damage at the site, For depth and frequency The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This is the static ionic conductivity. The vacuum permittivity, , , These are the weighting coefficients. The offset constant is used. This invention introduces a multi-band weighted fusion mechanism to achieve differentiated control of different dielectric characteristics, such as polarization, relaxation, and conductivity, significantly improving the model's ability to identify the degree of damage. By establishing a mapping model between tissue dielectric parameters and the degree of damage, the amplitude-frequency characteristics of the complex dielectric spectrum are directly correlated with the degree of damage. By combining depth-dependent relaxation time and distribution coefficient, the changes in dielectric behavior caused by variations in tissue microstructure are effectively captured, providing a sensitive indicator for the quantitative detection of early micro-damage.

[0016] Optionally, establishing the damage determination function for the pelvic tissue under test based on the mapping relationship includes: establishing the damage determination function for the pelvic tissue under test based on the mapping relationship between the dielectric parameters of the tissue at different depths and the degree of damage, wherein the damage determination function is as follows:

[0017] ,

[0018] in, The damage determination value is the damage determination value of the damage determination function. Indicates the first Characteristic frequencies and multiple damage state variables The dielectric constant of the following is given. Indicates the first Characteristic frequencies and multiple damage state variables The conductivity at that point Indicates the first Characteristic frequencies and multiple damage state variables The loss factor below, , , Weighting coefficients representing frequency and damage dependence. , Indicates the multi-damage compensation coefficient. Indicates tissue density, Indicates moisture content. , Represents the coefficients of the quadratic term that depend on frequency. Indicates the offset terms related to damage. Indicates the number of characteristic frequency points. , It is a non-linear correction exponent. The coefficient of the damage interaction term. For the first Characteristic functions of damage For the first This invention provides characteristic functions for various types of pelvic injuries. By integrating the relationship between multi-band dielectric parameters and dynamic injury variables, it achieves precise quantitative assessment of pelvic tissue injuries. By introducing an adaptive compensation mechanism and offset term for injury status, it significantly improves the model's adaptability to different pathological stages and individual differences, enhancing its clinical applicability. By constructing frequency-related quadratic terms and nonlinear correction terms, it effectively suppresses interference from high-frequency measurement noise, improving the stability and reliability of injury assessment. Through the design of injury interaction terms and the collaborative calculation of multiple injury characteristic functions, it achieves collaborative diagnosis of complex pelvic injuries, solving the problem of existing methods' insensitivity to concurrent pathological detection.

[0019] Optionally, obtaining the damage grade and damage type determination results of the pelvic tissue under test using the damage determination function includes: obtaining a damage determination value using the damage determination function; determining the damage grade using the damage determination value; determining a damage feature function based on the damage grade, the damage feature function including a temperature damage feature function, a pressure damage feature function, and an ischemic damage feature function; and obtaining the determination results of single damage type and complex damage type of the pelvic tissue under test according to the damage feature function. This invention achieves objective grading of pelvic tissue damage by dynamically matching the damage determination value with multi-dimensional damage grade standards; significantly improves the accuracy of identifying single-type damage by establishing specific damage feature functions for temperature, pressure, and ischemia; and achieves accurate analysis of complex pelvic injuries through nonlinear combination analysis of damage feature functions, solving the technical problem of distinguishing mixed pathological types.

[0020] Optionally, obtaining the determination results of single injury type and complex injury type of the pelvic tissue under test based on the injury feature function includes: determining the interactive contributions of temperature and pressure, temperature and ischemia, and pressure and ischemia based on the injury feature function; determining the dominant injury type based on the interactive contributions; and obtaining the determination results of single injury type and complex injury type of the pelvic tissue under test through the dominant injury type. This invention achieves precise quantification of each injury component in complex injuries by analyzing the interactive contributions of temperature, pressure, and ischemia injury feature functions, overcoming the limitation of existing methods in analyzing mixed injuries; and significantly improves the accuracy of distinguishing between single and complex injuries by identifying the dominant injury type.

[0021] Optionally, obtaining the determination results of single injury type and complex injury type of the pelvic tissue under test based on the injury feature function includes: determining the interactive contributions of temperature and pressure, temperature and ischemia, and pressure and ischemia based on the injury feature function; determining the dominant injury type based on the interactive contributions; and obtaining the determination results of single injury type and complex injury type of the pelvic tissue under test through the dominant injury type. This invention achieves precise quantification of the contribution of each injury component in complex pelvic injuries by establishing a dynamic interactive model of temperature, pressure, and ischemia injuries; significantly improves the accuracy of distinguishing between single and complex injuries by analyzing the interactive contributions of various injury features and identifying the dominant injury factor; achieves adaptive discrimination of different injury types by constructing a dynamic weight allocation mechanism for injury interactive contributions, significantly improving the clinical applicability of injury diagnosis; and forms a diagnostic chain from single injury identification to complex injury analysis by integrating a collaborative analysis algorithm of multiple injury features, providing a reliable basis for the formulation of precision medicine plans.

[0022] Secondly, the present invention provides a deep tissue damage detection system based on dielectric spectrum analysis, comprising an input device, a processor, an output device, and a memory, wherein the input device, the processor, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprising program instructions, the processor being configured to call the program instructions, and the system using the aforementioned deep tissue damage detection method based on dielectric spectrum analysis. The system provided by this invention has a high degree of system integration and smooth information transmission between its components. Through deep coupling modeling of multi-band dielectric parameters and tissue damage characteristics, it achieves accurate and non-invasive detection of deep tissue damage, overcoming the limitations of existing invasive biopsies and imaging examinations. By establishing a dynamic damage state adaptive dielectric parameter analysis model, it significantly improves the system's sensitivity and specificity in identifying tissue damage of different depths and types. Through an intelligent analysis algorithm that integrates dielectric spectrum characteristics and multi-physics field parameters, it achieves quantitative assessment of complex deep tissue damage, providing a new dimension of objective evidence for clinical diagnosis. By linking dielectric parameters with histopathology, it forms a fully automated analysis process from data acquisition to damage assessment, greatly improving clinical detection efficiency and diagnostic consistency. Attached Figure Description

[0023] Figure 1 This is a flowchart of a deep tissue damage detection method based on dielectric spectrum analysis according to an embodiment of the present invention;

[0024] Figure 2 This is a flowchart illustrating the determination of damage level and damage type according to an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of a deep tissue damage detection system based on dielectric spectrum analysis according to an embodiment of the present invention. Detailed Implementation

[0026] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0027] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0028] Please see Figure 1 The present invention provides a method for detecting deep tissue damage based on dielectric spectrum analysis, the method comprising the following steps:

[0029] S1. Obtain the complex permittivity spectrum of the pelvic tissue to be tested.

[0030] In one embodiment, a vector network analyzer is first used as the signal source to generate a swept-frequency electromagnetic wave signal with a frequency range of 1MHz-10GHz. Simultaneously, the transmitting antenna is placed on one side of the surface of the pelvic tissue to be tested, ensuring good coupling between the antenna and the tissue surface to reduce signal loss. It should be noted that the pelvic tissue is a deep tissue located in the lower abdominal cavity, containing reproductive, urinary, and some digestive organs, and is situated relatively deep within the abdominal cavity.

[0031] Furthermore, a receiving antenna is placed on the other side of the pelvic tissue to be tested to receive transmitted signals; at the same time, a reflected signal receiving device is set up near the transmitting antenna to acquire reflected signals.

[0032] It is important to note that when electromagnetic wave signals are emitted from the transmitting antenna and pass through the pelvic tissue being tested, they will inevitably first pass through the surface tissue. This surface tissue will reflect, scatter, and absorb the signal, resulting in the received transmitted and reflected signals containing information about the surface tissue. If the signal from the surface tissue is not separated, it is impossible to accurately obtain the signal characteristics of the pelvic tissue itself being tested.

[0033] Furthermore, based on the transmitted signal and the reflected signal, a time-frequency joint sparse reconstruction algorithm is designed. This algorithm reconstructs the original signal by combining information from the time and frequency domains and utilizing the sparsity of the signal. In this invention, the algorithm constructs and solves an optimization problem to separate the pelvic tissue signal and the superficial tissue signal from the mixed signal of the transmitted and reflected signals. For this algorithm, the following optimization problem is constructed:

[0034] ,

[0035] in, This represents the received signal vector, which includes both transmitted and reflected signals. The observation matrix describes how the signal is observed during the measurement process, mapping the original signal space to the observed signal space. The sparse transform matrix transforms the signal from the time domain to the frequency domain, making the signal exhibit sparsity in the frequency domain. , This represents the regularization parameter, used to control the performance of the reconstructed signal in terms of fidelity, sparsity, and smoothness. Let be the sparse signal vector to be reconstructed. , which are elements in the sparse signal vector. , These represent the index numbers of the row vector and column vector, respectively. The denominator is a very small positive number to avoid a zero denominator. It's important to note that the core of this algorithm is to leverage the sparsity of the signal in the time-frequency domain to construct a composite regularized optimization problem, and then solve this problem to achieve signal separation.

[0036] Furthermore, the sparse signal vector is obtained by utilizing the aforementioned time-frequency joint sparse reconstruction algorithm. Then, using a time-frequency mask, the signals of the pelvic tissue under test were separated. and surface tissue signals The time-frequency mask is a key separation technique in signal processing, used to extract target components from mixed signals. In the problem of separating pelvic signals and superficial tissue signals, the mask's role is to distinguish the different distribution characteristics of the two types of signals in the time-frequency domain.

[0037] Further, an inverse sparse transform is performed to obtain the separated pelvic tissue signals and surface tissue signals.

[0038] Specifically, let the signal of the separated pelvic tissue be... The signal of the surface tissue is , can be represented as:

[0039] ,

[0040] ,

[0041] in, Represents the sparse transformation matrix inverse transform, and Representing sparse signal vectors respectively The signals correspond to the pelvic tissue and superficial tissue being tested. Indicates time.

[0042] The advantage of this method lies in its ability to accurately distinguish the time-frequency characteristics of pelvic tissue and superficial tissue signals through time-frequency joint sparse reconstruction and mask separation, achieving high-precision signal separation. Specifically, the inverse sparse transform restores the original signal, effectively improving the signal-to-noise ratio and resolution, providing purer tissue feature information for medical diagnosis, and has significant clinical value.

[0043] Furthermore, a dielectric inversion model for deep generative adversarial networks is established.

[0044] Specifically, the time-domain signal and the frequency domain signal obtained through Fourier transform Concatenate them into a joint feature vector:

[0045] ,

[0046] in, For joint feature vectors, For the frequency domain amplitude spectrum, For frequency.

[0047] Furthermore, a deep generative adversarial network (GAN) structure is designed, which includes a generator and a discriminator.

[0048] Specifically, the generator uses a joint feature vector and random noise As input, output complex permittivity spectrum :

[0049]

[0050]

[0051] in, Let be the real part of the complex permittivity, representing the ability to store electrical energy. This represents the imaginary part of the complex permittivity, indicating the ability to dissipate electrical energy. Indicates organizational depth. For frequency, For generator parameters, The generator is a random noise that follows a Gaussian distribution. .

[0052] Furthermore, the discriminator takes as input real dielectric constant data or generated data, and outputs as a discrimination probability:

[0053] ,

[0054] in, For discriminator, The input data includes the actual dielectric constant data and the generated data. For discriminator parameters, To determine the probability.

[0055] Furthermore, a loss function is designed, which involves losses including generator loss and discriminator loss.

[0056] Specifically, the generator loss includes adversarial loss and physical constraint loss;

[0057] The generator loss includes adversarial loss and physical constraint loss;

[0058] ,

[0059] in, Indicates generator loss. Indicating resistance to loss, Represents physical constraint loss. , These are the weighting coefficients. The physical constraint loss is as follows:

[0060] ,

[0061] in, For electric field strength, For frequency, For depth, Satisfy the following expression:

[0062]

[0063] in, For wave number, At the speed of light, The spectrum represents the complex permittivity. The adversarial loss is expressed as the negative log-likelihood of the generated sample being judged as a real sample by the discriminator, i.e.:

[0064]

[0065] in, For noise vectors, For noise distribution, Samples generated by the generator The discriminator's judgment result on the generated samples. Let be the mathematical expectation.

[0066] Furthermore, the discriminator loss is as follows:

[0067] ,

[0068] in, For the loss of the discriminator, The dielectric constant data are from real samples. The discriminator's judgment result on the real sample data. For noise vectors, Enter information for additional conditions. For the discriminator to consider the sample It is the actual probability. Input the generator based on the conditions and noise vector The generated fake samples are used as input to the discriminator to train its ability to identify fake samples.

[0069] Furthermore, based on the above deep generative adversarial network (GAN) structure, considering the losses of the generator and discriminator, a deep GAN dielectric inversion model is established. This model includes a generator, a discriminator, generator loss, and discriminator loss. The deep GAN dielectric inversion model is a nonlinear inversion method based on adversarial learning, used to reconstruct the vertical distribution of the dielectric constant from observation data. Its core idea is to achieve high-precision, high-resolution dielectric parameter inversion through adversarial training of the generator and discriminator, combined with physical constraints.

[0070] Furthermore, to deepen Discretize by layer generator Output multidimensional tensors:

[0071] ,

[0072] Each layer corresponds to an independent physical constraint loss. , For frequency, For the first Layer depth, Enter information for additional conditions. For noise vectors, For generator parameters, The spectrum of complex permittivity at different depths is given.

[0073] Further, model training is performed, including the following steps:

[0074] Random initialization and ;

[0075] Iterative training: fixed ,renew , minimize ;fixed ,renew , minimize ;

[0076] Termination determination: When data is generated and real data error When the iteration stops, the iteration stops. This indicates the error threshold.

[0077] Furthermore, after training is complete, the generator It can directly predict any input signal Dielectric constant:

[0078] ,

[0079] Among them, noise Set to zero. These are the optimal parameters.

[0080] The advantage of this method is that it achieves high-precision and high-resolution dielectric parameter inversion through adversarial training between the generator and the discriminator, combined with hierarchical physical constraints. This overcomes the problems of existing methods that rely on linear approximation and have low computational efficiency. At the same time, through end-to-end prediction and hierarchical adaptive optimization, it significantly improves the accuracy and generalization ability of pelvic tissue vertical distribution reconstruction and reduces the dependence on a large amount of labeled data.

[0081] S2. Based on the spectrum of the complex permittivity, an improved bipolar Cole-Cole model is established.

[0082] In one embodiment, first for each depth The spectrum of complex permittivity Decompose and identify the existing relaxation processes, including low-frequency relaxation. and high frequency relaxation .

[0083] Furthermore, the relaxation time is extracted using the particle swarm optimization algorithm. and and its distribution coefficient and .

[0084] Furthermore, the analysis of relaxation time with depth Establish the pattern of change and A deep dependency model.

[0085] Specifically, The deep dependency model is as follows:

[0086] ,

[0087] The deep dependency model is as follows:

[0088] ,

[0089] in, , The scaling parameter is used to control the initial amplitude of the relaxation time. , To describe the decay rate of relaxation time as a function of depth, , This represents the offset, which indicates the baseline value of the relaxation time when the depth approaches infinity.

[0090] Furthermore, through the aforementioned and The deep dependency model is used to obtain the relaxation time distribution characteristics of deep dependencies.

[0091] Furthermore, based on the extracted relaxation time distribution characteristics, an improved bipolar Cole-Cole model is established, which satisfies the following expression:

[0092] ,

[0093] in, For different depths and different frequencies The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This is the static ionic conductivity. is the vacuum permittivity.

[0094] Furthermore, the model parameters were determined by fitting experimental data.

[0095] First, for deep tissue testing, corresponding samples were selected. Using a vector network analyzer equipped with a minimally invasive dielectric probe, time-domain voltage signals and frequency-domain amplitude spectra were collected layer by layer at 1mm depth intervals within the target frequency band of 1MHz-10GHz. This constructed a multi-depth, multi-frequency experimental dataset. Based on the particle swarm optimization algorithm, model parameters were first set according to the prior dielectric properties of normal / lesion tissues. , , , The initial search range is set, and the candidate parameter values ​​are substituted into the improved bipolar Cole-Cole model. The mean square error between the output complex permittivity and the measured data is calculated as the fitness. The particle position is adjusted iteratively until the mean square error is less than 0.01 or 1000 iterations are performed. Finally, 20% of the independently collected sample data are used for verification. If the correlation coefficient between the model's predicted complex permittivity and the measured value is greater than 0.95, the parameters are determined to be used as input feature constraints for the deep generative adversarial network (such as as initial conditions for the generator or real sample labels for the discriminator) to support the inversion of the complex permittivity. If the condition is not met, the sample stratification strategy or algorithm parameters are adjusted for refitting.

[0096] The improvements of this method are as follows: First, by introducing a depth function of relaxation time and dielectric constant, the existing homogeneous medium assumption is broken, and the dielectric constant variation of pelvic biological tissues is accurately characterized. Second, by separating high-frequency molecular polarization and low-frequency interface polarization through dual relaxation terms and combining dynamic distribution coefficients, the fitting accuracy of complex relaxation spectra is significantly improved. Third, by considering static conductivity, the distortion problem of existing models in the low-frequency region is solved, which is especially suitable for the detection of high water content tissues.

[0097] S3. Using the improved bipolar Cole-Cole model, establish a mapping relationship between tissue dielectric parameters and damage degree.

[0098] In one embodiment, using the improved bipolar Cole-Cole model, a mapping model between tissue dielectric parameters and damage degree is established, wherein the mapping model satisfies the following expression:

[0099] ,

[0100] in, For depth The degree of tissue damage at the site, For different depths and different frequencies The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This is the static ionic conductivity. The vacuum permittivity, , , These are the weighting coefficients. This is an offset constant used to adjust the baseline level of damage severity, ensuring that the model output is consistent with the observed data.

[0101] Furthermore, through the mapping model, the mapping relationship between the tissue dielectric parameters and the degree of damage is obtained, that is, the one-to-one correspondence between the tissue dielectric parameters and the degree of damage.

[0102] It should be noted that while the above mapping model is helpful for quickly screening damaged tissues, it does not address the questions of why the damage occurred and how to intervene.

[0103] The advantage of this method is that it achieves precise quantification of tissue damage through a mapping model between tissue dielectric parameters and damage degree, resulting in higher accuracy, adaptability, and real-time performance.

[0104] S4. Based on the mapping relationship, establish the damage determination function for the pelvic tissue to be tested.

[0105] To address the deficiencies in the mapping model in step S3, in one embodiment, a damage determination function for the pelvic tissue under test is established based on the mapping relationship between the dielectric parameters of tissue at different depths and the degree of damage. The damage determination function is as follows:

[0106] ,

[0107] in, The damage determination value is the result of the layered damage determination function. Indicates the first Characteristic frequencies and multiple damage state quantities The dielectric constant of the following is given. Indicates the first Characteristic frequencies and multiple damage state variables The conductivity at that point Indicates the first Characteristic frequencies and multiple damage state variables The loss factor below, , , Weighting coefficients representing frequency and damage dependence. , Indicates the multi-damage compensation coefficient. Indicates tissue density, Indicates moisture content. , Represents the coefficients of the quadratic term that depend on frequency. Indicates the offset terms related to damage. Indicates the number of characteristic frequency points. , It is a non-linear correction exponent. The coefficient of the damage interaction term. For the first Characteristic functions of damage For the first The characteristic function of a type of damage, the damage state variable This includes temperature, pressure, and mechanical strain.

[0108] The advantage of this method is that by establishing a multi-parameter, multi-frequency damage determination function, integrating dielectric properties, tissue parameters and multiple physical field variables, and introducing nonlinear correction and damage interaction terms, it significantly improves the accuracy and dynamic adaptability of pelvic tissue damage assessment and overcomes the shortcomings of existing models in analyzing complex damage states.

[0109] S5. Using the damage determination function, obtain the damage level and damage type determination results of the pelvic tissue to be tested.

[0110] Please see Figure 2 In one embodiment, the damage determination function is first used to obtain a damage determination value. .

[0111] Furthermore, the damage level is determined using the damage assessment value;

[0112] Specifically, the damage determination threshold is set, including a low threshold. Medium threshold High threshold ;

[0113] Furthermore, the damage determination value is compared with the threshold to obtain the following comparison result:

[0114] when At this time, it is considered a normal state, meaning that the pelvic tissues are basically intact and there are no obvious signs of damage.

[0115] when At this stage, it is classified as a mild injury, meaning that there is slight damage to the pelvic tissues, but the tissue function can still be maintained, and time is needed for maintenance.

[0116] when At this stage, it is classified as a moderate injury, meaning that the pelvic tissues are significantly damaged and their function is affected to some extent, requiring timely repair.

[0117] when At this stage, it is classified as severe damage, meaning that the pelvic tissues have been severely damaged and require extensive repair.

[0118] Furthermore, damage feature functions are extracted. Including temperature damage characteristic function Pressure damage characteristic function Ischemic injury characteristic function It satisfies the following expression:

[0119] ,

[0120] ,

[0121] ,

[0122] in, The current tissue temperature, This is normal physiological temperature. Due to the current pressure on the organization, The critical pressure that the tissue can tolerate. This represents the current blood perfusion volume. This represents normal blood perfusion.

[0123] Furthermore, the contribution of interaction terms is calculated, including:

[0124] Calculate the interaction contribution of temperature and pressure: ,in, The coefficient of the damage interaction term between temperature and pressure. The characteristic function of temperature damage, The characteristic function of pressure damage;

[0125] Calculate the interaction contribution between temperature and ischemia: ,in, This represents the coefficient of the interaction term between temperature and ischemia-induced damage. The characteristic function of temperature damage, The characteristic function of ischemic injury;

[0126] Calculate the interaction contribution between stress and ischemia: ,in, The coefficient of the interaction term between pressure and ischemia. The characteristic function of pressure damage, The characteristic function of ischemic injury;

[0127] Further, the dominant injury type is determined, which includes temperature injury, pressure injury, and ischemic injury. The characteristics of temperature injury include: Higher than normal value change, , Generates interactive contributions; the characteristics of the pressure damage include: rise, decline, Generates interactive contributions; the ischemic injury is characterized by: rise, change, It gets taller.

[0128] Further, single-damage assessment is performed. Specifically:

[0129] Set the damage interaction coefficient between temperature and pressure. Threshold of the interaction term coefficient between temperature and ischemia The interaction coefficient between stress and ischemia .

[0130] like and , This was determined to be temperature damage;

[0131] like and The injury was determined to be pressure damage.

[0132] like and Abnormal, diagnosed as ischemic injury.

[0133] in, , , They are respectively , , The threshold.

[0134] Further, a comprehensive damage assessment is performed. The details are as follows:

[0135] like , and If both exceed the corresponding threshold, it is determined to be a combined damage caused by temperature and pressure.

[0136] like , and If all values ​​are higher than the corresponding threshold, it is determined to be a combined injury of temperature and ischemia.

[0137] like , and If all values ​​are higher than the corresponding threshold, it is determined to be a combined injury of pressure and ischemia;

[0138] like , and All are higher than the corresponding threshold, and , and If the value is also higher than the corresponding threshold, it is determined to be a combined injury of temperature, pressure and ischemia.

[0139] The advantage of this method is that it achieves accurate quantitative assessment of pelvic tissue damage through multi-parameter hierarchical determination and interaction term analysis. It can distinguish between single-factor and complex damage types, and dynamically judge the damage level by combining physiological thresholds, providing a scientific basis for clinical diagnosis and repair. It is comprehensive, sensitive and operable.

[0140] Please see Figure 3 , Figure 3 This is a schematic diagram of the deep tissue damage detection system based on dielectric spectrum analysis in an embodiment of the present invention. The system includes an input device, a processor, an output device, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to call the program instructions, and the system uses the described deep tissue damage detection method based on dielectric spectrum analysis.

[0141] In this embodiment, the input device includes a swept-frequency electromagnetic wave transmitting module and a signal receiving module. The functions of the input device include: transmitting swept-frequency electromagnetic wave signals to the pelvic tissue under test to excite the tissue's dielectric response; simultaneously acquiring transmitted and reflected signals; and converting the original analog signal into a digital signal, which is then transmitted to a processor for subsequent analysis.

[0142] The processor includes a signal processing unit, a modeling and mapping unit, and a damage determination unit. The signal processing unit includes a time-frequency joint sparse reconstruction algorithm module and a deep generative adversarial network dielectric inversion module. The modeling and mapping unit includes an improved bipolar Cole-Cole model solver and a mapping model of dielectric parameters and damage severity. The damage determination unit includes a multi-parameter fusion damage determination function calculation engine and a damage type classifier. The processor's functions include: separating deep and superficial tissue signals using the time-frequency joint sparse reconstruction algorithm; inverting the complex dielectric constant spectrum of tissues at different depths using the model; establishing an improved bipolar Cole-Cole model and extracting depth-dependent relaxation times; calculating damage determination values, classifying damage levels, and analyzing the interactive contributions of temperature, pressure, and ischemic damage characteristic functions.

[0143] The output device includes a visual interactive interface and a report generation module. The functions of the output device include: real-time display of damage thermograms and key dielectric parameters at various depths of pelvic tissue; and generation of clinically readable damage detection reports, including analysis of damage grade, damage type, and damage cause.

[0144] The memory uses a high-speed solid-state drive, which features fast read and write speeds, large capacity, and high reliability. It is mainly used to store data input from input devices and the result data after processing by the processor, and can meet the needs of storing large amounts of data.

[0145] In summary, this invention achieves high-precision quantitative characterization of the dielectric properties of deep tissues by acquiring the complex dielectric constant spectrum of the pelvic tissue under test, providing a reliable data foundation for improving the sensitivity and specificity of damage detection. By establishing an improved bipolar Cole-Cole model, it overcomes the shortcomings of existing models in terms of insufficient fitting accuracy over a wide frequency range, providing a more accurate mathematical description of the dielectric behavior of pelvic tissues. By establishing a mapping relationship between tissue dielectric parameters and damage degree, it reveals the correlation between changes in dielectric properties and tissue pathological state, creating conditions for quantitative damage assessment. By constructing a damage judgment function, it achieves intelligent damage classification through multi-parameter fusion, providing an objective basis for distinguishing different damage levels and types. Through the determination results of damage level and damage type, it achieves non-invasive and rapid detection of deep tissue damage, providing a novel technical means for clinical diagnosis and treatment efficacy monitoring of pelvic diseases. This method effectively improves the sensitivity, specificity, and clinical applicability of deep tissue damage detection.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for detecting deep tissue damage based on dielectric spectrum analysis, characterized in that, The method includes the following steps: Obtain the complex permittivity spectrum of the pelvic tissue to be tested; Based on the aforementioned complex permittivity spectrum, an improved bipolar Cole-Cole model is established, including: Based on the spectrum of the complex permittivity, the depth-dependent relaxation time distribution characteristics are obtained; Based on the aforementioned relaxation time distribution characteristics, an improved bipolar Cole-Cole model is established; The improved bipolar Cole-Cole model satisfies the following expression: in, For different depths and different frequencies The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This refers to the static ionic conductivity. It is the vacuum permittivity; Using the improved bipolar Cole-Cole model, a mapping relationship between tissue dielectric parameters and damage degree is established; Based on the mapping relationship, a damage determination function for the pelvic tissue to be tested is established; Using the damage assessment function, the damage level and damage type of the pelvic tissue to be tested are determined. The determination results enable the detection of damage to deep tissues.

2. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 1, characterized in that, The acquisition of the complex permittivity spectrum of the pelvic tissue to be tested includes: A swept-frequency electromagnetic wave signal is emitted into the pelvic tissue to be tested to obtain the transmitted and reflected signals; Based on the transmitted signal and the reflected signal, a time-frequency joint sparse reconstruction algorithm is designed. The time-frequency joint sparse reconstruction algorithm is used to obtain the signal separation results of the pelvic tissue and surface tissue under test; Based on the signal separation results, a deep generative adversarial network dielectric inversion model is established; The complex permittivity spectrum of the pelvic tissue under test at different depths was obtained using the deep generative adversarial network dielectric inversion model.

3. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 1, characterized in that, The establishment of the mapping relationship between tissue dielectric parameters and damage degree using the improved bipolar Cole-Cole model includes: Using the improved bipolar Cole-Cole model, a mapping model between tissue dielectric parameters and damage degree is established; The mapping model is used to obtain the mapping relationship between the tissue dielectric parameters and the degree of damage.

4. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 3, characterized in that, The mapping model between the tissue dielectric parameters and the degree of damage satisfies the following expression: in, For depth The degree of tissue damage at the site, For depth and frequency The complex permittivity of the following, It is the high-frequency limiting dielectric constant. , For depth The relevant DC dielectric constant, , For depth Dependency relaxation time , The distribution coefficient, This refers to the static ionic conductivity. The vacuum permittivity, , , These are the weighting coefficients. This is the offset constant.

5. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 1, characterized in that, The damage determination function for the pelvic tissue to be tested, based on the mapping relationship, includes: Based on the mapping relationship between the dielectric parameters of tissue at different depths and the degree of damage, a damage determination function is established for the pelvic tissue to be tested. The damage determination function is as follows: in, The damage determination value is the damage determination value of the damage determination function. Indicates the first Characteristic frequencies and multiple damage state variables The dielectric constant of the following, Indicates the first Characteristic frequencies and multiple damage state variables The conductivity at that point Indicates the first Characteristic frequencies and multiple damage state variables The loss factor below, , , Weighting coefficients representing frequency and damage dependence. , Indicates the multi-damage compensation coefficient. Indicates tissue density, Indicates moisture content. , Represents the coefficients of the quadratic term that depend on frequency. Indicates the offset terms related to damage. Indicates the number of characteristic frequency points. , It is a non-linear correction exponent. The coefficient of the damage interaction term. For the first Characteristic functions of damage For the first Characteristic functions of a type of damage.

6. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 1, characterized in that, The step of using the damage assessment function to obtain the damage level and damage type of the pelvic tissue under test includes: The damage determination value is obtained using the damage determination function. The damage level is determined based on the damage assessment value. Based on the damage level, damage characteristic functions are determined, including temperature damage characteristic functions, pressure damage characteristic functions, and ischemic damage characteristic functions. Based on the damage feature function, the determination results of single damage type and complex damage type of the pelvic tissue to be tested are obtained.

7. The method for detecting deep tissue damage based on dielectric spectrum analysis according to claim 6, characterized in that, The step of obtaining the determination results of single injury type and complex injury type of the pelvic tissue to be tested based on the injury feature function includes: Based on the injury characteristic function, determine the interactive contributions of temperature and pressure, temperature and ischemia, and pressure and ischemia. Based on the interaction contributions, the dominant damage type is determined; Based on the dominant injury type, the determination results of single injury type and complex injury type of the pelvic tissue under test are obtained.

8. A deep tissue damage detection system based on dielectric spectrum analysis, wherein the system uses the deep tissue damage detection method based on dielectric spectrum analysis according to any one of claims 1 to 7, characterized in that, The system includes an input device, a processor, an output device, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions, and the processor is configured to invoke the program instructions.

Images