Target disturbance analysis method, device, medical equipment and readable storage medium

Abstract

The present application relates to a target disturbance analysis method, device, medical equipment and readable storage medium, which obtain target radio frequency channel state information corresponding to target biological tissue; determine the target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information through a trained target disturbance analysis model; based on the target radio frequency channel state information corresponding to the target biological tissue, the bioelectromagnetic characteristics of the target biological tissue can be more comprehensively reflected, laying a foundation for improving the accuracy of the target disturbance analysis; and then based on the trained target disturbance analysis model, according to the target radio frequency channel state information, the target disturbance analysis result corresponding to the target biological tissue can be accurately determined, thereby improving the accuracy of the target disturbance analysis result.

Description

Technical Field

[0001] The present application relates to the field of bio-electromagnetic technology, and in particular to a target disturbance analysis method, device, medical equipment and readable storage medium. Background Art

[0002] As an interdisciplinary field, bioelectromagnetism studies the phenomena of electricity, magnetism and their interactions in biological tissues, covering the behavior of excitable tissues, the distribution of current and potential inside organisms, changes in magnetic fields inside and outside the body, and the response mechanism of biological cells to electromagnetic fields. There are various charged ions and biomacromolecules in organisms, which will generate bioelectric currents under the action of external electromagnetic fields or membrane potentials. At the same time, ion channels and potential differences on cell membranes are the basis of cellular electrical activity, and different tissues and organs have different electrical and magnetic conductivities. These differences bring challenges to tissue position and state detection technology based on bioelectromagnetic principles. In related technologies, magnetic induction tomography equipment is often used to realize the analysis of target disturbances in biological tissues.

[0003] However, the biological tissue target perturbation analysis method in the related art has the problem of low accuracy. Summary of the invention

[0004] Based on this, it is necessary to provide a target disturbance analysis method, device, medical equipment and readable storage medium to address the above technical problems.

[0005] In a first aspect, the present application provides a target disturbance analysis method, the method comprising:

[0006] Acquiring target radio frequency channel state information corresponding to the target biological tissue;

[0007] Determining a target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information by using the trained target disturbance analysis model;

[0008] The method for constructing the trained target disturbance analysis model includes:

[0009] Acquire a radio frequency channel state information group corresponding to the bionic sample group; the radio frequency channel state information group includes a plurality of radio frequency channel state information; the radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers;

[0010] Performing data preprocessing on each piece of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each piece of the radio frequency channel state information;

[0011] Constructing a corresponding bionic sample data set according to the frequency domain feature information corresponding to the plurality of radio frequency channel state information;

[0012] The target disturbance analysis model is trained according to the bionic sample data set to obtain a trained target disturbance analysis model.

[0013] In one embodiment, the bionic sample group includes a first category sample group, a second category sample group, and a third category sample group, and the step of obtaining a radio frequency channel state information group corresponding to the bionic sample group includes:

[0014] Acquiring first radio frequency channel state information corresponding to each sample in a first category sample group; the multiple samples in the first category sample group have different conductivities but the same volumes and relative positions;

[0015] Acquire second radio frequency channel state information corresponding to each sample in the second category sample group; the multiple samples in the second category sample group have different volumes but the same conductivity and relative positions;

[0016] Acquire third radio frequency channel state information corresponding to each sample in a third category sample group; the plurality of samples in the third category sample group have different relative positions but the same volume and conductivity;

[0017] A radio frequency channel state information group corresponding to the bionic sample group is obtained according to the plurality of first radio frequency channel state information, the plurality of second radio frequency channel state information and the plurality of third radio frequency channel state information.

[0018] In one of the embodiments, performing data preprocessing on each of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each of the radio frequency channel state information includes:

[0019] Performing feature extraction on each of the radio frequency channel state information in the radio frequency channel state information group to obtain original frequency domain feature information corresponding to each of the radio frequency channel state information;

[0020] The original frequency domain characteristic information is denoised to obtain frequency domain characteristic information corresponding to each radio frequency channel state information.

[0021] In one embodiment, the original frequency domain feature information includes original phase information, and the frequency domain feature information includes a phase signal; and the denoising process is performed on the original frequency domain feature information to obtain the frequency domain feature information corresponding to each of the radio frequency channel state information, including:

[0022] Performing phase recovery processing on the original phase information to obtain continuous phase information corresponding to each of the radio frequency channel state information;

[0023] The continuous phase information is linearly transformed to obtain phase information corresponding to each radio frequency channel state information.

[0024] In one embodiment, constructing a corresponding bionic sample data set according to the frequency domain feature information corresponding to the radio frequency channel state information includes:

[0025] Constructing a first sample data set according to frequency domain feature information corresponding to a plurality of first radio frequency channel state information;

[0026] Constructing a second sample data set according to frequency domain feature information corresponding to the plurality of second radio frequency channel state information;

[0027] constructing a third sample data set according to frequency domain feature information corresponding to multiple third radio frequency channel state information;

[0028] A bionic sample data set is obtained according to the first sample data set, the second sample data set and the third sample data set.

[0029] In one embodiment, the target disturbance analysis model is trained according to the bionic sample data set to obtain a trained target disturbance analysis model, including:

[0030] Based on the first sample data set, the second sample data set and the third sample data set, a target disturbance analysis model is trained to obtain a trained target disturbance analysis model;

[0031] The output of the trained target disturbance analysis model includes conductivity category results, volume category results, and relative position category results.

[0032] In a second aspect, the present application further provides a target disturbance analysis device, the device comprising:

[0033] An acquisition module, used to acquire target radio frequency channel state information corresponding to the target biological tissue;

[0034] An analysis module, configured to determine a target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information through a trained target disturbance analysis model;

[0035] The analysis module includes an information acquisition module, a preprocessing module, a data set construction module, and a model training module;

[0036] The information acquisition module is used to acquire a radio frequency channel state information group corresponding to the bionic sample group; the radio frequency channel state information group includes a plurality of radio frequency channel state information; the radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers;

[0037] The preprocessing module is used to perform data preprocessing on each of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each of the radio frequency channel state information;

[0038] The data set construction module is used to construct a corresponding bionic sample data set according to the frequency domain feature information corresponding to the plurality of radio frequency channel state information;

[0039] The model training module is used to train the target disturbance analysis model according to the bionic sample data set to obtain a trained target disturbance analysis model.

[0040] In a third aspect, the present application further provides a medical device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any one of the embodiments of the first aspect are implemented.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in any one of the embodiments of the first aspect above are implemented.

[0042] In a fifth aspect, the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method described in any one of the embodiments of the first aspect above.

[0043] The above-mentioned target perturbation analysis method, device, medical equipment and readable storage medium; when performing target perturbation analysis, based on the target RF channel state information corresponding to the target biological tissue, it can more comprehensively reflect the bioelectromagnetic characteristics of the target biological tissue, laying a foundation for improving the accuracy of target perturbation analysis; then based on the trained target perturbation analysis model, according to the target RF channel state information, it can accurately determine the target perturbation analysis result corresponding to the target biological tissue, thereby improving the accuracy of the target perturbation analysis result. In terms of constructing the trained target perturbation analysis model, based on the RF channel state information group corresponding to the bionic sample group, a corresponding bionic sample data set is constructed; then, according to the bionic sample data set, the target perturbation analysis model is trained to obtain a trained target perturbation analysis model, thereby enhancing the generalization ability and accuracy of the target perturbation analysis model, thereby laying a foundation for improving the accuracy of the target perturbation analysis result. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 is an application environment diagram of a target disturbance analysis method in an embodiment;

[0046] Figure 2 Schematic diagram of a target disturbance analysis method in one embodiment;

[0047] Figure 3 A schematic diagram of a flow chart of steps for constructing a target disturbance analysis model after training in one embodiment;

[0048] Figure 4 A schematic diagram of a process for obtaining radio frequency channel status information in an embodiment;

[0049] Figure 5 A schematic flow chart of the steps of obtaining a radio frequency channel state information group corresponding to a bionic sample group in one embodiment;

[0050] Figure 6 A schematic diagram of a data structure of radio frequency channel state information in a specific embodiment;

[0051] Figure 7 is a schematic diagram of a radio frequency channel target disturbance analysis system in a specific embodiment;

[0052] Figure 8 is a structural block diagram of a target disturbance analysis device in one embodiment;

[0053] Fig. 9 FIG. 1 is a diagram showing the internal structure of a medical device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The target disturbance analysis method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0056] In an exemplary embodiment, Figure 2 As shown in the figure, a target disturbance analysis method is provided, which is applied to Figure 1 The terminal in the example is used for explanation. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0057] Step S201, obtaining target radio frequency channel state information corresponding to the target biological tissue.

[0058] Among them, the target biological tissue can be human tissue, cells, organs, etc.; the target RF channel state information includes channel state signals corresponding to multiple subcarriers; the target RF channel state information is used to describe the changes of multiple subcarriers after passing through the target biological tissue.

[0059] It should be noted that based on orthogonal frequency division multiplexing technology, the RF signal can be decomposed into multiple lower frequency subcarriers; each subcarrier is an independent carrier signal that can transmit information simultaneously at different frequencies; and each subcarrier is orthogonal in frequency, which means that they can transmit data simultaneously within the same total bandwidth without interfering with each other. Based on this, by analyzing the target RF channel state information, the bio-electromagnetic properties of the target biological tissue can be accurately and comprehensively understood.

[0060] Optionally, in an exemplary embodiment, the method for obtaining the target radio frequency channel state information corresponding to the target biological tissue may be: using the radio frequency channel target disturbance analysis system to collect the target radio frequency channel state information corresponding to the target biological tissue. The radio frequency channel target disturbance analysis system includes a control module and a signal transceiver module; the control module includes a notebook; the signal transceiver module includes an antenna; the antenna is an integrated transceiver, that is, it can both transmit and receive radio frequency channel state information; the control module and the signal transceiver module are connected by a connecting line.

[0061] It should be noted that the model, quantity and other specifications of the laptops, antennas and connecting cables need to be selected according to the actual collection requirements and are not specifically limited here; for example, the control module includes two laptops equipped with 5300 network cards (half-height), and the laptops meet the 3.13 kernel version of Ubuntu14.06; the signal transceiver module includes six 20cm antennas, and the antenna specifications are 30MHz and 50MHz dual-band inner needle 10DBI; the two laptops and the six antennas are connected by six 0.2m connecting cables, and the connecting cable model is IPEX to SMA outer screw inner hole.

[0062] Step S202: determining a target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information through the trained target disturbance analysis model.

[0063] Among them, the target disturbance analysis results may include but are not limited to conductivity category results, volume category results and relative position category results of the target biological tissue; among them, conductivity can reflect the ability of the target biological tissue to conduct current; volume can reflect the spatial size of the target biological tissue; relative position can reflect the position of the target biological tissue relative to other tissues.

[0064] Exemplarily, the target RF channel state information is input into a trained target disturbance analysis model, and through the trained target disturbance analysis model, the target disturbance analysis results corresponding to the target biological tissue can be output, that is, the conductivity category results, volume category results and relative position category results corresponding to the target biological tissue.

[0065] Among them, Figure 3 As shown, the method for constructing the trained target disturbance analysis model includes the following steps:

[0066] Step S301, obtaining a radio frequency channel state information group corresponding to a bionic sample group.

[0067] Among them, the bionic sample group is used to simulate biological tissues in different states; the simulation sample group includes multiple sample groups, each sample group includes multiple samples; wherein the sample is a salt solution; it should be noted that different sample groups represent biological tissues corresponding to different test conditions; based on the control variable method, the parameters corresponding to the test conditions are taken as variables, and a single variable is changed to obtain the corresponding sample group; for example, if the test condition is conductivity, conductivity needs to be taken as a variable, and other parameters such as volume, relative position, etc. remain unchanged to obtain multiple samples, and then obtain the corresponding conductivity sample group.

[0068] The RF channel state information group includes multiple RF channel state information; the RF channel state information includes channel state information corresponding to multiple subcarriers; the channel state information refers to the characteristic data of the subcarrier after passing through the channel, such as amplitude, phase, etc.; the channel is the medium or path for signal transmission, responsible for transmitting information from the transmitter to the receiver, such as an antenna. Exemplarily, the process of obtaining RF channel state information, i.e., channel state information corresponding to multiple subcarriers based on orthogonal frequency division multiplexing technology is shown in the figure below: Figure 4 shown.

[0069] It should be noted that when the sample is close to the subcarrier transmission channel, the conductivity and other characteristics of the sample will affect the channel state information corresponding to the subcarrier, such as phase and amplitude. Therefore, by analyzing the channel state information corresponding to multiple subcarriers, the specific characteristics corresponding to the sample can be inferred.

[0070] It should be noted that the method for obtaining the RF channel state information group corresponding to the bionic sample group may be, but is not limited to, collecting the RF channel state information group corresponding to the bionic sample group using a RF channel target disturbance analysis system. The RF channel target disturbance analysis system is the same as the RF channel target disturbance analysis system described in the above embodiment, and will not be described in detail here.

[0071] Step S302: perform data preprocessing on each piece of radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each piece of radio frequency channel state information.

[0072] The data preprocessing method may include, but is not limited to, feature extraction and denoising. The frequency domain feature information includes amplitude information and phase information corresponding to multiple subcarriers; the amplitude information is used to characterize the strength or amplitude of the channel state information corresponding to different subcarriers; the phase information is used to characterize the phase offset of the channel state information corresponding to different subcarriers.

[0073] Step S303: construct a corresponding bionic sample data set according to the frequency domain feature information corresponding to the multiple radio frequency channel state information.

[0074] Exemplarily, for each sample group in multiple sample groups in the bionic sample group, the RF channel state information corresponding to each sample in the sample group is obtained, and data preprocessing is performed on the RF channel state information of each sample to obtain frequency domain feature information corresponding to the RF channel state information, that is, amplitude information and phase information corresponding to multiple subcarriers; according to the sample characteristics, the corresponding frequency domain feature information is labeled to obtain labeled frequency domain feature information; according to the labeled frequency domain feature information corresponding to each of the multiple samples in each sample group, a corresponding bionic sample data set is constructed.

[0075] Optionally, according to the sample characteristics, the corresponding frequency domain feature information is labeled, and the method for obtaining the labeled frequency domain feature information may be: according to the preset classification rules, according to the actual conductivity, actual volume and actual relative position of the sample, the conductivity category, volume category and relative position category corresponding to the sample are determined; according to the conductivity category, volume category and relative position category corresponding to the sample, the frequency domain feature information corresponding to the sample is labeled to obtain the labeled frequency domain feature information. Among them, the preset classification rules need to be set according to the actual classification requirements, and are not specifically limited here.

[0076] Step S304: training the target disturbance analysis model according to the bionic sample data set to obtain a trained target disturbance analysis model.

[0077] Among them, the target disturbance analysis model is used to solve the multi-classification problem to obtain the target disturbance analysis results, and the target disturbance analysis results include conductivity category results, volume category results and relative position category results.

[0078] Optionally, the target perturbation analysis model can be but is not limited to a support vector machine classification model; in an exemplary embodiment, the target perturbation analysis model is a support vector machine classification model, and the internal structure of the model includes feature space mapping, decision hyperplane, support vector and multi-class classification strategy; wherein, feature space mapping is to map the input data to a high-dimensional feature space through a kernel function (the default is a linear kernel) so as to find a linearly separable hyperplane in the high-dimensional space; the decision hyperplane is a linear equation used to find an optimal segmentation surface in the feature space so as to separate data of different categories and maximize the interval between two categories of data; the support vectors are the sample points closest to the decision hyperplane, which define the location of the decision boundary, and these points play a key role in the optimization process; the multi-class classification strategy refers to the use of an error correction output coding strategy to decompose the multi-class classification problem into multiple binary classification problems for the multi-class classification problem, and to process the multi-class classification task by combining multiple binary classifiers.

[0079] In this embodiment, based on the target radio frequency channel state information corresponding to the target biological tissue, the bioelectromagnetic characteristics of the target biological tissue can be more comprehensively reflected, laying a foundation for improving the accuracy of the target perturbation analysis; then based on the trained target perturbation analysis model, according to the target radio frequency channel state information, the target perturbation analysis result corresponding to the target biological tissue can be accurately determined, thereby improving the accuracy of the target perturbation analysis result. In terms of constructing the trained target perturbation analysis model, based on the radio frequency channel state information group corresponding to the bionic sample group, a corresponding bionic sample data set is constructed; then, based on the bionic sample data set, the target perturbation analysis model is trained to obtain a trained target perturbation analysis model, thereby enhancing the generalization ability and accuracy of the target perturbation analysis model, thereby laying a foundation for improving the accuracy of the target perturbation analysis result.

[0080] In one embodiment, the bionic sample group includes a first category sample group, a second category sample group, and a third category sample group. It should be noted that the first category sample group, the second category sample group, and the third category sample group are sample groups corresponding to different test conditions. Figure 5 As shown, obtaining the radio frequency channel state information group corresponding to the bionic sample group includes the following steps:

[0081] Step S501: Acquire first radio frequency channel state information corresponding to each sample in a first category sample group.

[0082] The multiple samples in the first category sample group have different conductivities but the same volumes and relative positions. The conductivity can reflect the ability of the sample to conduct current; the volume can reflect the spatial size of the sample; and the relative position can reflect the position of the sample relative to the background solution. The sample is a salt solution.

[0083] Exemplarily, multiple salt solutions are configured according to the preset conductivity to ensure that the multiple salt solutions have different conductivities but the same volume and the same relative position in the background solution. The preset conductivity needs to be set according to actual needs and is not specifically limited here. For example, the preset conductivity is 0.2S / m, 0.8S / m, 1.0S / m, and 1.5S / m. The background solution is a saturated calcium sulfate solution, which is placed in a circular water tank mold with a diameter of 16 cm and a mold height of 5.5 cm. The liquid level is 3.3 cm high and the volume is 660 ml.

[0084] It is understandable that when obtaining the first RF channel state information corresponding to each sample in the first category sample group, it is necessary to ensure that the relative position of each sample in the background solution remains the same while ensuring that the conductivity of multiple samples is different and the volume is the same. Based on this, based on the RF channel target disturbance analysis system, a multi-band RF signal with a preset frequency is used. According to the preset sampling frequency and the preset time range, the first RF channel state information corresponding to each sample in the first category sample group can be collected. Among them, the preset frequency, preset sampling frequency and preset time range need to be set according to the actual collection requirements, and are not specifically limited here. For example, the preset frequency is 30MHz, the preset sampling frequency is 100Hz, and the preset time range is 30s.

[0085] It should be noted that the RF channel target disturbance analysis system includes 270 channels; the 270 channels are specifically represented by 9 antenna link teams formed by 3 transmitting antennas and 3 receiving antennas, and each antenna link team contains channel state information corresponding to 30 subcarriers. Therefore, the RF channel target disturbance analysis system includes 270 channels.

[0086] Step S502: Acquire second radio frequency channel state information corresponding to each sample in the second category sample group.

[0087] The multiple samples in the second category sample group have different volumes but the same conductivity and relative positions.

[0088] Exemplarily, multiple salt solutions are configured according to preset volumes to ensure that the volumes of the multiple salt solutions are different but the conductivity is the same (for example, all are 1.63S / m), and the relative positions in the background solution are the same. The preset volume needs to be set according to actual needs and is not specifically limited here. For example, the preset volume is 5ml, 10ml, 15ml, 20ml, 25ml, 30ml, 35ml, 40ml, 45ml, 50ml, 55ml, and 60ml.

[0089] It can be understood that when obtaining the second RF channel state information corresponding to each sample in the second category sample group, it is necessary to ensure that the relative position of each sample in the background solution remains the same while ensuring that the volumes of multiple samples are different and the conductivity is the same. Based on this, based on the RF channel target disturbance analysis system, a multi-band RF signal with a preset frequency is used. According to the preset sampling frequency and the preset time range, the second RF channel state information corresponding to each sample in the second category sample group can be collected.

[0090] Step S503: Acquire third radio frequency channel state information corresponding to each sample in the third category sample group.

[0091] The multiple samples in the third category sample group have different relative positions but the same volume and conductivity.

[0092] It can be understood that when obtaining the third RF channel state information corresponding to each sample in the third category sample group, it is necessary to ensure that the relative position of each sample in the background solution remains different while ensuring that the conductivity and volume of multiple samples are the same. Based on this, based on the RF channel target disturbance analysis system, a multi-band RF signal with a preset frequency is used. According to the preset sampling frequency and the preset time range, the third RF channel state information corresponding to each sample in the third category sample group can be collected.

[0093] Step S504: obtaining a radio frequency channel state information group corresponding to the bionic sample group according to the plurality of first radio frequency channel state information, the plurality of second radio frequency channel state information and the plurality of third radio frequency channel state information.

[0094] In this embodiment, the first category sample group, the second category sample group and the third category sample group are configured based on the control variable method, which lays a foundation for determining multiple first RF channel state information, multiple second RF channel state information and multiple third RF channel state information; and then based on the multiple first RF channel state information, multiple second RF channel state information and multiple third RF channel state information, the RF channel state information group corresponding to the bionic sample group is obtained, which lays a data foundation for solving the multi-classification problem.

[0095] In one embodiment, data preprocessing is performed on each radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each radio frequency channel state information, including the following steps:

[0096] Step 1: extract features from each radio frequency channel state information in the radio frequency channel state information group to obtain original frequency domain feature information corresponding to each radio frequency channel state information.

[0097] The original frequency domain feature information includes original phase information.

[0098] The method for extracting features from each RF channel state information in the RF channel state information group may be: extracting frequency domain features from each RF channel state information in the RF channel state information group; wherein the RF channel state information may be represented by formula (1).

[0099] (1)

[0100] in, , represents the amplitude information corresponding to the RF channel state information, , represents the original phase information corresponding to the RF channel state information, that is, the original phase information corresponding to the i-th subcarrier.

[0101] According to formula (1), the amplitude information corresponding to the RF channel state information It can be expressed by formula (2); the original phase information corresponding to the RF channel state information It can be expressed by formula (3).

[0102] (2)

[0103] (3)

[0104] Among them, since the radio frequency channel state information can be expressed in a complex form, , represents the real part of the radio frequency channel state information; , represents the imaginary part of the RF channel state information.

[0105] Step 2: De-noising the original frequency domain feature information to obtain frequency domain feature information corresponding to each radio frequency channel state information.

[0106] The frequency domain feature information includes a phase signal.

[0107] It should be noted that due to the original phase information contains a lot of noise, as shown in formula (4), so the original phase information Perform denoising.

[0108] (4)

[0109] in, , represents the real phase information; , represents the timing error offset (or delay), which is mainly caused by factors such as multipath effect, oscillator phase noise, signal processing delay and hardware delay; , represents the random phase offset caused by the RF channel target perturbation analysis system; , represents the random noise in the measurement process; It is the frequency index of the subcarrier, which is used to indicate the position and number of the subcarrier; N is the number of FFT points, N = 64.

[0110] Preferably, step 2 performs denoising on the original frequency domain feature information to obtain frequency domain feature information corresponding to each radio frequency channel state information, including the following steps:

[0111] Step 2.1, performing phase recovery processing on the original phase information to obtain continuous phase information corresponding to each RF channel state information.

[0112] For example, since a phase information jump at π will cause phase distortion, it is necessary to perform phase recovery processing on the original phase information to obtain continuous phase information corresponding to each RF channel state information to ensure the continuity of the phase information. The principle of phase recovery processing is shown in formula (5).

[0113] (5)

[0114] in, , represents the difference between the true value and the measured value of the phase at the nth time sampling point; , which represents the difference between the true value and the measured value of the phase at the n-1th time sampling point.

[0115] It can be understood that performing phase recovery processing on the original phase information can effectively remove the timing error offset And the random phase offset caused by the RF channel target perturbation analysis system .

[0116] Step 2.2, linear transformation is performed on the continuous phase information to obtain phase information corresponding to each RF channel state information.

[0117] For example, the timing error offset This causes the phase of the subcarrier to change linearly with the frequency of the subcarrier. The random phase offset caused by the RF channel target disturbance analysis system , which is usually related to the system hardware and is a static error that remains constant during the entire transmission. And the random phase offset caused by the RF channel target perturbation analysis system To avoid the impact caused by this, it is necessary to perform linear transformation on the continuous phase information based on the phase change rate a calculated by formula (6) and the average value of the phase information of all subcarriers calculated by formula (7) to obtain the phase information corresponding to each RF channel state information.

[0118] (6)

[0119] (7)

[0120] Where a represents the phase change rate of the subcarrier spacing, and b represents the average value of the phase information of all subcarriers; , represents the original phase information; n, represents the number of subcarriers; and , respectively represent the frequency index of the 1st subcarrier and the frequency index of the nth subcarrier.

[0121] It can be understood that linear transformation of continuous phase information can further remove the timing error offset. and random noise in the measurement process .

[0122] In this embodiment, by performing feature extraction on each RF channel state information in the RF channel state information group, the original frequency domain feature information corresponding to each RF channel state information can be accurately obtained, and then the original frequency domain feature information is subjected to phase recovery processing and linear transformation processing, so as to eliminate the noise contained in the original frequency domain feature information and obtain the frequency domain feature information corresponding to each RF channel state information, thereby laying a foundation for improving the accuracy of the model.

[0123] In one embodiment, constructing a corresponding bionic sample data set according to frequency domain feature information corresponding to radio frequency channel state information includes the following steps:

[0124] Step 1: construct a first sample data set according to frequency domain feature information corresponding to multiple first radio frequency channel state information.

[0125] Step 2: construct a second sample data set according to frequency domain feature information corresponding to multiple second radio frequency channel state information.

[0126] Step 3: construct a third sample data set according to frequency domain feature information corresponding to multiple third radio frequency channel state information.

[0127] Step 4: Obtain a bionic sample data set according to the first sample data set, the second sample data set and the third sample data set.

[0128] It should be noted that the method for determining the frequency domain characteristic information corresponding to multiple first RF channel state information, the frequency domain characteristic information corresponding to multiple second RF channel state information, and the frequency domain characteristic information corresponding to multiple second RF channel state information is the same as that described in the above embodiment and will not be repeated here.

[0129] It is understandable that in order to solve the multi-classification problem, it is necessary to construct corresponding data sets for each category, wherein the first sample data set corresponds to the conductivity category, the second sample data set corresponds to the volume category, and the third sample data set corresponds to the phase position category.

[0130] Exemplarily, according to the frequency domain feature information corresponding to multiple first radio frequency channel state information, the method of constructing the first sample data set may be: according to the characteristics of each sample in the first category sample group, labeling the corresponding frequency domain feature information to obtain labeled frequency domain feature information; and constructing the first sample data set according to the labeled frequency domain feature information.

[0131] Optionally, according to the characteristics of each sample in the first category sample group, the corresponding frequency domain feature information is labeled, and the method for obtaining the labeled frequency domain feature information can be: according to the preset classification rules, according to the actual conductivity, actual volume and actual relative position of the corresponding sample in the first category sample group, the conductivity category, volume category and relative position category corresponding to the sample are determined; according to the conductivity category, volume category and relative position category corresponding to the sample, the frequency domain feature information corresponding to the sample is labeled to obtain the labeled frequency domain feature information. Among them, the preset classification rules need to be set according to the actual classification requirements, and are not specifically limited here.

[0132] In an exemplary embodiment, if the amount of subcarrier data is 30, the labeled frequency domain feature information corresponding to a sample in the first category sample group is: ;in, is the amplitude information corresponding to the i-th subcarrier, is the phase information corresponding to the i-th subcarrier; , is the corresponding label, including the volume category, conductivity category and relative position category corresponding to the sample, for example It can be: [volume category: category 2; conductivity category: category 5; relative position category: category 3]; and then according to the labeled frequency domain feature information corresponding to the multiple samples in the first category sample group, the corresponding first sample data set can be constructed. Similarly, the construction principle of the second sample data set and the third sample data set is the same as the construction principle of the first sample data set, which will not be repeated here.

[0133] In this embodiment, based on the frequency domain feature information corresponding to multiple first RF channel state information, the frequency domain feature information corresponding to multiple second RF channel state information, and the frequency domain feature information corresponding to multiple third RF channel state information, the corresponding first sample data set, second sample data set, and third sample data set can be constructed respectively, laying a data foundation for achieving accurate prediction of the volume, conductivity, and phase position of the target biological tissue.

[0134] In one embodiment, training a target disturbance analysis model according to a bionic sample data set to obtain a trained target disturbance analysis model includes the following steps:

[0135] Based on the first sample data set, the second sample data set and the third sample data set, the target disturbance analysis model is trained to obtain a trained target disturbance analysis model.

[0136] Among them, the output of the trained target disturbance analysis model includes conductivity category results, volume category results and relative position category results.

[0137] Exemplarily, before training the target disturbance analysis model, it is necessary to divide the first sample data set, the second sample data set, and the third sample data set respectively based on the preset data set division rule to obtain the training set and test set corresponding to each data set. It should be noted that the preset data set division rule needs to be set according to actual needs and is not specifically limited here. For example, the preset data set division rule is to take every 10 consecutive samples as an independent subset, and select the samples in the order of 1, 5, and 10 in the independent subset as the test set according to the ratio of 7:3 for training set: test set, and the rest as training sets.

[0138] It can be understood that based on the above-mentioned preset data set division rules, it can ensure that the samples corresponding to the test set are more representative, that is, they can well represent the distribution of the entire data set, and at the same time ensure that the test set and the training set have similar statistical characteristics, so that the performance of the model on the test set is more valuable for reference, avoiding deviation or overfitting of the model in the evaluation stage due to uneven sample selection.

[0139] It should be noted that the effectiveness of the model can be evaluated by calculating the consistency between the predicted values ​​of the test set and the actual labels.

[0140] In this embodiment, the target perturbation analysis model is trained based on the first sample data set, the second sample data set and the third sample data set to obtain a trained target perturbation analysis model, thereby improving the accuracy and generalization ability of the trained target perturbation analysis model, and laying the foundation for realizing the conductivity prediction, volume prediction and relative position prediction corresponding to the target biological tissue.

[0141] In a specific embodiment, the radio frequency channel target disturbance analysis system recorded in the above embodiment is a multiple input multiple output (MIMO) system, and the data structure diagram of the radio frequency channel state information is as follows: Figure 6 As shown, the RF channel state information is represented by a three-dimensional matrix; the horizontal axis is the transmitting antenna, which represents multiple transmitting antennas in the RF channel target disturbance analysis system. Assuming there are N transmitting antennas, each transmitting antenna will transmit different RF signals at a subcarrier (frequency) and time t. The vertical axis is the receiving antenna, which represents multiple receiving antennas in the RF channel target disturbance analysis system. Assuming there are M receiving antennas, the RF signal received by each receiving antenna contains different paths and interference signals from each transmitting antenna; the depth axis is the subcarrier, which represents multiple subcarriers in the RF channel target disturbance analysis system; each subcarrier corresponds to a different RF frequency; assuming there are S subcarriers, the RF channel response of each subcarrier at different time points t may be different. Represents the channel state information from the i-th (i=1,2,…,N) transmitting antenna to the j-th (j=1,2,…,M) receiving antenna at the S-th subcarrier and the t-th time.

[0142] In a specific embodiment, a schematic diagram of a radio frequency channel target disturbance analysis system is shown in FIG. Figure 7 It should be noted that Figure 7 What is given is a schematic diagram of the radio frequency channel target disturbance analysis system when collecting radio frequency channel state information corresponding to each sample in the bionic sample group; Figure 7 In the figure, 701 is an antenna, 702 is a background solution, and 703 is a sample. It can be understood that the antenna 701 is a transmitter-receiver integrated.

[0143] The above-mentioned target perturbation analysis method, when performing target perturbation analysis, can more comprehensively reflect the bioelectromagnetic characteristics of the target biological tissue based on the target radio frequency channel state information corresponding to the target biological tissue, laying a foundation for improving the accuracy of the target perturbation analysis; then based on the trained target perturbation analysis model, according to the target radio frequency channel state information, the target perturbation analysis result corresponding to the target biological tissue can be accurately determined, thereby improving the accuracy of the target perturbation analysis result. In terms of constructing the trained target perturbation analysis model, based on the radio frequency channel state information group corresponding to the bionic sample group, a corresponding bionic sample data set is constructed; then, according to the bionic sample data set, the target perturbation analysis model is trained to obtain the trained target perturbation analysis model, thereby enhancing the generalization ability and accuracy of the target perturbation analysis model, thereby laying a foundation for improving the accuracy of the target perturbation analysis result.

[0144] It should be understood that, although the steps in the flowcharts of the embodiments labeled above are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the steps or stages in other steps.

[0145] Based on the same inventive concept, the embodiment of the present application also provides a target disturbance analysis device for implementing the target disturbance analysis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more target disturbance analysis device embodiments provided below can refer to the limitations on the target disturbance analysis method above, and will not be repeated here.

[0146] In an exemplary embodiment, Figure 8 As shown, a target disturbance analysis device is provided, including: an acquisition module 801 and an analysis module 802;

[0147] An acquisition module 801 is used to acquire target radio frequency channel state information corresponding to a target biological tissue;

[0148] An analysis module 802 is used to determine a target disturbance analysis result corresponding to a target biological tissue according to the target radio frequency channel state information through the trained target disturbance analysis model;

[0149] The analysis module 802 includes an information acquisition module, a preprocessing module, a data set construction module, and a model training module;

[0150] An information acquisition module is used to acquire a radio frequency channel state information group corresponding to the bionic sample group; the radio frequency channel state information group includes a plurality of radio frequency channel state information; the radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers;

[0151] A preprocessing module, used to perform data preprocessing on each radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each radio frequency channel state information;

[0152] A data set construction module, used to construct a corresponding bionic sample data set according to frequency domain feature information corresponding to multiple radio frequency channel state information;

[0153] The model training module is used to train the target disturbance analysis model according to the bionic sample data set to obtain a trained target disturbance analysis model.

[0154] The above-mentioned target disturbance analysis device, based on the target radio frequency channel state information corresponding to the target biological tissue, can more comprehensively reflect the bioelectromagnetic characteristics of the target biological tissue, laying a foundation for improving the accuracy of the target disturbance analysis; further, based on the trained target disturbance analysis model, according to the target radio frequency channel state information, the target disturbance analysis result corresponding to the target biological tissue can be accurately determined, thereby improving the accuracy of the target disturbance analysis result. In terms of constructing the trained target disturbance analysis model, based on the radio frequency channel state information group corresponding to the bionic sample group, a corresponding bionic sample data set is constructed; further, based on the bionic sample data set, the target disturbance analysis model is trained to obtain the trained target disturbance analysis model, thereby enhancing the generalization ability and accuracy of the target disturbance analysis model, thereby laying a foundation for improving the accuracy of the target disturbance analysis result.

[0155] In one embodiment, the bionic sample group includes a first category sample group, a second category sample group and a third category sample group, and the information acquisition module is further used to

[0156] Acquire first radio frequency channel state information corresponding to each sample in the first category sample group; the multiple samples in the first category sample group have different conductivities but the same volumes and relative positions;

[0157] Acquire second radio frequency channel state information corresponding to each sample in the second category sample group; the multiple samples in the second category sample group have different volumes but the same conductivity and relative positions;

[0158] Acquiring third radio frequency channel state information corresponding to each sample in the third category sample group; the plurality of samples in the third category sample group have different relative positions but the same volume and conductivity;

[0159] According to the plurality of first radio frequency channel state information, the plurality of second radio frequency channel state information and the plurality of third radio frequency channel state information, a radio frequency channel state information group corresponding to the bionic sample group is obtained.

[0160] In one embodiment, the preprocessing module is further used to perform data preprocessing on each radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each radio frequency channel state information, including:

[0161] Extracting features of each radio frequency channel state information in the radio frequency channel state information group to obtain original frequency domain feature information corresponding to each radio frequency channel state information;

[0162] The original frequency domain characteristic information is denoised to obtain the frequency domain characteristic information corresponding to each RF channel state information.

[0163] In one embodiment, the original frequency domain feature information includes the original phase information, and the frequency domain feature information includes the phase signal; the preprocessing module is also used to

[0164] Performing phase recovery processing on the original phase information to obtain continuous phase information corresponding to each RF channel state information;

[0165] The continuous phase information is linearly transformed to obtain the phase information corresponding to each radio frequency channel state information.

[0166] In one embodiment, the data set construction module is also used to

[0167] Constructing a first sample data set according to frequency domain feature information corresponding to a plurality of first radio frequency channel state information;

[0168] Constructing a second sample data set according to frequency domain feature information corresponding to the plurality of second radio frequency channel state information;

[0169] constructing a third sample data set according to frequency domain feature information corresponding to multiple third radio frequency channel state information;

[0170] A bionic sample data set is obtained according to the first sample data set, the second sample data set and the third sample data set.

[0171] In one embodiment, the model training module is used to train the target disturbance analysis model based on the first sample data set, the second sample data set and the third sample data set to obtain a trained target disturbance analysis model;

[0172] The outputs of the trained target disturbance analysis model include conductivity category results, volume category results, and relative position category results.

[0173] Each module in the above-mentioned target disturbance analysis device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0174] In an exemplary embodiment, a medical device is provided. The medical device may be a server, and its internal structure diagram may be as follows: Fig. 9As shown. The medical device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the medical device is used to provide computing and control capabilities. The memory of the medical device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the medical device is used to store target disturbance analysis related data. The input / output interface of the medical device is used to exchange information between the processor and an external device. The communication interface of the medical device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a target disturbance analysis method is implemented.

[0175] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0176] In one embodiment, a medical device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0177] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0178] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0179] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0180] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0181] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A target disturbance analysis method, characterized in that: The method comprises: Acquire target radio frequency channel state information corresponding to the target biological tissue; the target radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers; the target radio frequency channel state information is used to describe changes in the plurality of subcarriers after passing through the target biological tissue; Determine the target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information through the trained target disturbance analysis model; the target disturbance analysis result includes the conductivity category result, volume category result and relative position category result of the target biological tissue; The method for constructing the trained target disturbance analysis model includes: Acquire a radio frequency channel state information group corresponding to the bionic sample group; the radio frequency channel state information group includes a plurality of radio frequency channel state information; the radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers; Performing data preprocessing on each of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain characteristic information corresponding to each of the radio frequency channel state information; the frequency domain characteristic information includes amplitude information and phase information corresponding to a plurality of subcarriers; Constructing a corresponding bionic sample data set according to the frequency domain feature information corresponding to the plurality of radio frequency channel state information; According to the bionic sample data set, a target disturbance analysis model is trained to obtain a trained target disturbance analysis model; The bionic sample group includes a first category sample group, a second category sample group and a third category sample group, and the step of obtaining a radio frequency channel state information group corresponding to the bionic sample group includes: Acquire the first radio frequency channel state information corresponding to each sample in the first category sample group; the multiple samples in the first category sample group have different conductivity, the same volume and relative position; the sample is a salt solution; the relative position refers to the position of the sample relative to the background solution; the background solution is a saturated calcium sulfate solution; the sample is built into the background solution, and the sample is isolated from the background solution; the volume refers to the spatial size of the sample; Acquire second radio frequency channel state information corresponding to each sample in the second category sample group; the multiple samples in the second category sample group have different volumes but the same conductivity and relative positions; Acquire third radio frequency channel state information corresponding to each sample in a third category sample group; the plurality of samples in the third category sample group have different relative positions but the same volume and conductivity; A radio frequency channel state information group corresponding to the bionic sample group is obtained according to the plurality of first radio frequency channel state information, the plurality of second radio frequency channel state information and the plurality of third radio frequency channel state information.

2. The method according to claim 1, characterized in that The performing data preprocessing on each of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain feature information corresponding to each of the radio frequency channel state information includes: Performing feature extraction on each of the radio frequency channel state information in the radio frequency channel state information group to obtain original frequency domain feature information corresponding to each of the radio frequency channel state information; The original frequency domain characteristic information is denoised to obtain frequency domain characteristic information corresponding to each radio frequency channel state information.

3. The method according to claim 2, characterized in that The original frequency domain feature information includes original phase information, and the frequency domain feature information includes a phase signal; and the denoising process is performed on the original frequency domain feature information to obtain the frequency domain feature information corresponding to each of the radio frequency channel state information, including: Performing phase recovery processing on the original phase information to obtain continuous phase information corresponding to each of the radio frequency channel state information; The continuous phase information is linearly transformed to obtain phase information corresponding to each radio frequency channel state information.

4. The method according to claim 1, characterized in that The step of constructing a corresponding bionic sample data set according to the frequency domain feature information corresponding to the radio frequency channel state information includes: Constructing a first sample data set according to frequency domain feature information corresponding to a plurality of first radio frequency channel state information; Constructing a second sample data set according to frequency domain feature information corresponding to the plurality of second radio frequency channel state information; constructing a third sample data set according to frequency domain feature information corresponding to multiple third radio frequency channel state information; A bionic sample data set is obtained according to the first sample data set, the second sample data set and the third sample data set.

5. The method according to claim 4, characterized in that The target disturbance analysis model is trained according to the bionic sample data set to obtain a trained target disturbance analysis model, including: Based on the first sample data set, the second sample data set and the third sample data set, a target disturbance analysis model is trained to obtain a trained target disturbance analysis model; The output of the trained target disturbance analysis model includes conductivity category results, volume category results, and relative position category results.

6. A target disturbance analysis device, characterized in that: The device comprises: An acquisition module, used to acquire target radio frequency channel state information corresponding to the target biological tissue; the target radio frequency channel state information includes channel state information corresponding to multiple subcarriers; the target radio frequency channel state information is used to describe the change of multiple subcarriers after passing through the target biological tissue; An analysis module, used to determine the target disturbance analysis result corresponding to the target biological tissue according to the target radio frequency channel state information through the trained target disturbance analysis model; the target disturbance analysis result includes the conductivity category result, volume category result and relative position category result of the target biological tissue; The analysis module includes an information acquisition module, a preprocessing module, a data set construction module, and a model training module; The information acquisition module is used to acquire a radio frequency channel state information group corresponding to the bionic sample group; the radio frequency channel state information group includes a plurality of radio frequency channel state information; the radio frequency channel state information includes channel state information corresponding to a plurality of subcarriers; The preprocessing module is used to perform data preprocessing on each of the radio frequency channel state information in the radio frequency channel state information group to obtain frequency domain characteristic information corresponding to each of the radio frequency channel state information; the frequency domain characteristic information includes amplitude information and phase information corresponding to multiple subcarriers; The data set construction module is used to construct a corresponding bionic sample data set according to the frequency domain feature information corresponding to the plurality of radio frequency channel state information; The model training module is used to train the target disturbance analysis model according to the bionic sample data set to obtain a trained target disturbance analysis model; The bionic sample group includes a first category sample group, a second category sample group and a third category sample group, and the information acquisition module is further used to obtain the first radio frequency channel state information corresponding to each sample in the first category sample group; the multiple samples in the first category sample group have different conductivity, the same volume and relative position; the sample is a salt solution; the relative position refers to the position of the sample relative to the background solution; the background solution is a saturated calcium sulfate solution; the sample is built into the background solution, and the sample is isolated from the background solution; the volume refers to the spatial size of the sample; Acquire second radio frequency channel state information corresponding to each sample in the second category sample group; the multiple samples in the second category sample group have different volumes but the same conductivity and relative positions; Acquire third radio frequency channel state information corresponding to each sample in a third category sample group; the plurality of samples in the third category sample group have different relative positions but the same volume and conductivity; A radio frequency channel state information group corresponding to the bionic sample group is obtained according to the plurality of first radio frequency channel state information, the plurality of second radio frequency channel state information and the plurality of third radio frequency channel state information.

7. A medical device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images