Tissue identification method based on multi-frequency thermoacoustic imaging combined feature extraction

Abstract

The invention discloses a method for realizing component and content identification of different tissues through multi-frequency thermoacoustic imaging combined feature extraction, and belongs to the field of microwave thermoacoustic imaging. The method comprises the following steps: radiating multi-frequency microwaves to biological tissues, receiving a thermoacoustic signal by using an ultrasonic transducer, and amplifying the signal for image reconstruction. The multi-frequency microwaves are pulse microwaves with various different frequencies, different frequencies are switched during imaging to excite the tested tissue in sequence, multiple groups of thermoacoustic images excited by microwaves with different frequencies are correspondingly obtained, and then the multiple groups of thermoacoustic images are fused to obtain a multi-frequency thermoacoustic image. Feature extraction is carried out by selecting an interested area in a single-frequency image, and distribution of different tissue components is identified and contents of the different tissue components are calculated based on a difference algorithm. The invention further provides a system for implementing the method. According to the method, clinical actual requirements are closely combined, efficient identification of different tissue components and accurate contents is realized, and the method has a very good application prospect.

Description

Technical Field

[0001] The present invention belongs to the field of microwave thermoacoustic imaging and photoacoustic imaging, comprises a system and a method for calculating the content of different tissues, and in particular relates to a method for tissue identification based on multi-frequency thermoacoustic imaging combined with feature extraction. Background Art

[0002] Medical imaging plays a vital role in the field of clinical diagnosis. The microwave thermoacoustic imaging system is a non-invasive biomedical imaging method that combines the advantages of high contrast of microwave imaging and high resolution of ultrasound imaging. The principle of microwave thermoacoustic imaging technology can be summarized as follows: First, pulsed microwaves are used as excitation sources to irradiate biological tissues. Second, biological tissues absorb microwave energy and then expand, generating mechanical waves (or thermoacoustic signals), which radiate outward. After receiving the signal, the probe amplifies the signal. The acquisition card collects the signal. Finally, the image is reconstructed using relevant algorithms. According to the type of electromagnetic pulse, microwave thermoacoustic imaging is generally divided into photoacoustic imaging and microwave-induced thermoacoustic imaging. Both can achieve high-resolution and high-contrast imaging. However, due to the high frequency of the photoacoustic excitation source in photoacoustic imaging, methods related to photoacoustic imaging are usually focused on superficial tissues. On the other hand, microwave thermoacoustic imaging technology has potential advantages in detecting deep tissues, which has attracted the attention of many researchers. Microwave thermoacoustic imaging technology continues to develop, and research in application fields continues to expand, showing broad prospects and unique value in clinical and application.

[0003] However, in traditional microwave thermoacoustic imaging technology, imaging mainly relies on single-frequency microwave excitation. Although the imaging effect is relatively significant for tissues with strong microwave absorption ability, it presents less obvious imaging characteristics for biological tissues with weak microwave absorption performance. This phenomenon of reduced microwave absorption may be due to two main reasons: one is that the microwave absorption coefficient of the tissue itself is low; the other is that the size of the tissue structure is too small, so that it is smaller than the tomographic projection thickness used by the detector, which leads to the weakening of the overall absorption effect of the entire tissue under the action of microwaves, and is affected by the surrounding larger-sized tissues, which ultimately leads to poor imaging quality.

[0004] Traditional single-frequency microwave thermoacoustic imaging technology has achieved visualization of specific biological tissue structures to a certain extent, but due to its limitations, when facing different tissues with different microwave absorption characteristics, especially those with weak microwave absorption or mixed with multiple tissues, such as fat, muscle and blood vessels, the image resolution and quantitative analysis capabilities under single-frequency excitation are obviously limited. Existing technologies are difficult to accurately distinguish and quantify the relative content of these different tissue components, thus limiting the in-depth application of this technology in clinical diagnosis and scientific research.

[0005] In 2023, Chi Zihui et al. (Science China: Technological Sciences, 2024, 54(2):333-350) proposed a dual-frequency microwave thermoacoustic imaging method and application research. This method effectively reduces the influence of bones on imaging quality by using two short-pulse microwaves with different center frequencies to excite tissues, combined with energy calibration and weighted subtraction techniques. This technology improves the imaging capability of biological tissues with complex bone structures, and contributes to the development of traditional microwave thermoacoustic imaging technology based on single-frequency excitation. However, the main contribution of this technology is to improve the imaging quality of the bone area, and does not involve the identification of different tissue components and their contents. At present, there is a lack of relevant research, and it is impossible to analyze the data of dual-frequency microwave thermoacoustic imaging through appropriate data processing methods to achieve the identification of different tissue components and their contents. In short, although the existing technology has been improved in some aspects, it still has many shortcomings, especially in the identification and content analysis of multiple tissue components. Summary of the invention

[0006] In order to overcome the limitations of single-frequency microwave thermoacoustic imaging in the prior art in tissue component identification and content assessment, the present invention provides a multi-frequency thermoacoustic imaging joint feature extraction method, which aims to achieve efficient identification of different tissue components and precise contents. This method is closely combined with actual clinical needs, and evaluates different tissue components and contents through a multi-frequency excitation mechanism, verifies the accuracy of microwave thermoacoustics in evaluating the content in different tissues, verifies the feasibility and stability of microwave thermoacoustics in evaluating tissue content, and improves the sensitivity and reliability of detecting multiple tissue components.

[0007] A multi-frequency thermoacoustic imaging system, comprising:

[0008] An image acquisition system is configured to acquire thermoacoustic signals of a sample under at least two frequency excitations, reconstruct a single-frequency thermoacoustic image through an imaging algorithm, and obtain a multi-frequency thermoacoustic image through a fusion algorithm;

[0009] The image analysis module is configured to calculate the content of different tissues in the sample through tissue thermoacoustic signals under at least two frequency excitations.

[0010] Preferably, the tissue is selected from fat, muscle or blood vessel.

[0011] Preferably, the multi-frequency thermoacoustic image is a dual-frequency thermoacoustic image, and the calculation formula of tissue content is:

[0012]

[0013] Among them, F and M refer to two different tissues in the sample, TA_F is the proportion of tissue F in the sample, and C F and C M are the contents of the two tissues, σ m1and σ m2 is the conductivity of tissue M at frequencies f1 and f2, σ f1 and σ f2 is the conductivity of tissue F at frequencies f1 and f2, E1 and E2 are the electric field strengths at frequencies f1 and f2, Γ f and Γ m are the Gruneisen parameters of tissue F and tissue M respectively, and A1 and A2 are the thermoacoustic signal intensity information at frequencies f1 and f2 respectively.

[0014] Preferably, the at least two frequencies are at least two frequencies having different microwave center frequencies but the same other parameters, and the other parameters include pulse width, pulse power and pulse repetition frequency.

[0015] Preferably, the imaging algorithm is selected from a back-projection algorithm, a delay and superposition algorithm, a filtered back-projection algorithm, a time reversal algorithm or compressed sensing.

[0016] Preferably, the fusion algorithm obtains a multi-frequency thermoacoustic image through single-frequency thermoacoustic data or images, specifically including: fusing multiple groups of single-frequency thermoacoustic images or directly fusing multiple groups of single-frequency thermoacoustic data, the fusion adopts differential fusion, the initial fusion weight factor is determined by the theoretical microwave absorption coefficient, after obtaining the multi-frequency thermoacoustic image, the smoothness and focus intensity of the tissue area to be highlighted are judged, if the tissue is smooth and focused, the multi-frequency thermoacoustic image obtained after the fusion is directly output and accompanied by several groups of single-frequency images, if the tissue is not smooth and not focused, the weight factor is adjusted until the tissue is smooth and focused, and then a group of multi-frequency thermoacoustic images is output and accompanied by multiple groups of single-frequency thermoacoustic images.

[0017] Preferably, the image acquisition system comprises: a multi-frequency microwave source, a coaxial line, an antenna, an ultrasonic transducer, an amplifier, a multi-channel data acquisition card and a computer.

[0018] The multi-frequency microwave source is used to generate pulse microwaves of different frequencies, which are transmitted to the antenna via a coaxial cable, and the microwaves are radiated to the sample under test. The sample under test absorbs microwave energy to generate ultrasonic waves, i.e., thermoacoustic signals. The thermoacoustic signals under microwave excitation of different frequencies are detected by ultrasonic transducers respectively, and the detected signals are amplified by amplifiers. The amplified signals are collected by multi-channel data acquisition cards, thereby obtaining thermoacoustic data under excitation of different frequencies. The entire process is controlled by a computer, and a single-frequency thermoacoustic image is reconstructed using an imaging algorithm according to the relationship between time domain information and spatial position. Then, a multi-frequency thermoacoustic image is obtained by using a fusion algorithm for the single-frequency thermoacoustic data or image. The frequency in the multi-frequency microwave thermoacoustic imaging refers to the microwave center frequency. The single-frequency thermoacoustic image is used to analyze biological tissues with strong single-frequency microwave energy absorption, and the area of ​​interest in the single-frequency image is selected to calculate the content of different tissue components. The multi-frequency thermoacoustic image is used to highlight tissues with weak single-frequency microwave energy absorption and large microwave energy absorption changes with frequency changes, and to detect corresponding lesions.

[0019] Preferably, the pulse microwaves of different frequencies generated by the multi-frequency microwave source are generated by the same variable-frequency microwave source, or by multiple single-frequency microwave sources; the pulse microwaves of different frequencies need to maintain the same pulse width and peak power; or the multi-frequency microwave source has a power monitoring and real-time feedback function;

[0020] And / or, the antenna is a non-frequency-variable antenna covering multiple frequency ranges or for multiple required radiation frequency points; or, the antenna is a wide-band antenna covering multiple frequency ranges; the antenna needs to ensure that the field distribution of multiple frequency points used for excitation imaging is consistent; the antenna with consistent field distribution excites the sample under test;

[0021] And / or, the ultrasonic transducer, amplifier, and multi-channel data acquisition card constitute a thermoacoustic signal acquisition module, and when the multi-frequency microwave excitation is switched, the ultrasonic transducer, amplifier, multi-channel data acquisition card, antenna, and biological tissue under test remain stationary;

[0022] And / or, the multi-frequency microwave source and the antenna constitute a microwave excitation module, wherein the multi-frequency microwave source is composed of multiple independent microwave sources that can output the same pulse width, the same repetition frequency, the same power and different frequencies, or the multi-frequency microwave source is a microwave source with a variable frequency but ensuring that other pulse microwave parameters remain unchanged, and the other pulse microwave parameters include pulse width, pulse power and pulse repetition frequency.

[0023] The present invention also provides a multi-frequency microwave thermoacoustic imaging method based on the above system, comprising the following steps:

[0024] Step 1: When the multi-frequency is dual-frequency, for biological tissues with weak microwave absorption under single-frequency microwave excitation, a frequency range in which the theoretical microwave absorption coefficient changes greatly but the microwave absorption coefficient of surrounding tissues changes less is selected, and two boundary frequency points of this frequency range are selected as two excitation frequency points for dual-frequency microwave thermoacoustic imaging;

[0025] Step 2: Turn on the device and set parameters for initialization;

[0026] Step 3: Place the sample to be tested into the coupling liquid;

[0027] Step 4: Use a computer to trigger the multi-frequency microwave source to generate a pulse microwave with a center frequency, and trigger the data acquisition card to start working: a pulse microwave with a center frequency is radiated to the biological tissue to be tested through the antenna, and the tissue generates a thermoacoustic signal based on the thermoacoustic effect. The thermoacoustic signal is detected by the ultrasound detector, converted into an electrical signal by the ultrasound detector, and collected by the data acquisition card, and then stored in the computer and processed by single-frequency thermoacoustic image reconstruction;

[0028] Step 5: Keep all experimental conditions unchanged;

[0029] Step 6: Use a computer to trigger the multi-frequency microwave source to generate another pulse microwave with a center frequency, and trigger the data acquisition card to start working: another pulse microwave pulse with a center frequency is radiated to the biological tissue to be tested through the antenna, and the tissue generates ultrasonic waves, i.e., thermoacoustic signals, based on the thermoacoustic effect. The thermoacoustic signals are detected by the ultrasonic detector, converted into electrical signals by the ultrasonic detector, and collected by the data acquisition card, and then stored in the computer and processed by single-frequency thermoacoustic image reconstruction, that is, the thermoacoustic imaging of this single-frequency excitation is completed; according to actual needs, more than two frequency microwaves can be generated to excite thermoacoustic signals;

[0030] Step 7: Use the fusion algorithm to post-process the completed single-frequency thermoacoustic image, or directly process the single-frequency thermoacoustic data stored in the computer;

[0031] Step 8: Export images, where the single-frequency thermoacoustic image is used to analyze biological tissues with strong single-frequency microwave energy absorption and detect corresponding lesions, while the multi-frequency thermoacoustic image is used to highlight tissues with weak single-frequency microwave energy absorption but large changes in microwave energy absorption with frequency changes and detect corresponding lesions.

[0032] The present invention also provides a method for calculating tissue content based on the above system, comprising the following steps:

[0033] Step A: collecting the electric field intensity of a reference object that is insensitive to the change of microwave center frequency at each excitation frequency as the normalized base of the multi-frequency microwave thermoacoustic signal;

[0034] Step B: acquiring multiple single-frequency thermoacoustic images;

[0035] Step C: calculating the normalized electric field intensity of the single-frequency thermoacoustic image;

[0036] Step D: Extract tissue features from the single-frequency thermoacoustic image to obtain the area where the tissue of interest is located.

[0037] Step E: Calculate the content of different tissues based on the selected area where the tissue of interest is located.

[0038] The technical solution of the present invention is as follows: The method for evaluating the content of different tissues using the multi-frequency microwave thermoacoustic imaging technology comprises the following steps:

[0039] 1. System construction: Design and construct a multi-frequency microwave thermoacoustic imaging system, which includes a multi-frequency microwave source, antenna, ultrasonic transducer, amplifier and multiplexer, multi-channel data acquisition card and computer control system.

[0040] 2. Microwave generation and transmission: A multi-frequency microwave source generates a series of microwave signals with different center frequencies and pulses, which are transmitted to the antenna in sequence through a coaxial cable.

[0041] 3. Microwave radiation and thermoacoustic signal generation: The antenna radiates the received microwaves of different frequencies to the biological tissue under test. After the tissue absorbs the microwave energy, it is converted into an ultrasonic signal, which is a thermoacoustic signal.

[0042] 4. Thermoacoustic signal detection and amplification: The array ultrasonic detector captures the thermoacoustic signals generated under different frequency excitations in turn and sends them to the array amplifier circuit for signal enhancement processing.

[0043] 5. Signal acquisition and recording: The amplified thermoacoustic signal is collected and stored in real time by a multi-channel data acquisition card to form multiple sets of thermoacoustic data sets at different frequencies.

[0044] 6. Image reconstruction: Based on the time domain information and spatial position relationship of the thermoacoustic signal, imaging algorithms such as back-projection algorithm, time reversal algorithm or compressed sensing are used to reconstruct the image of each set of single-frequency thermoacoustic data to generate multiple single-frequency thermoacoustic images.

[0045] 7. Image fusion and analysis: The obtained single-frequency thermoacoustic images are integrated using a fusion algorithm to obtain high-resolution and high-contrast multi-frequency thermoacoustic images.

[0046] 8. Calculation of tissue content:

[0047] ① Calculation of normalized electric field intensity: Based on the characteristics of the thermoacoustic signal at each frequency, the normalized electric field intensity at different frequencies is first calculated on the single-frequency image.

[0048] ② Feature extraction: Extract tissue features from single-frequency thermoacoustic images, including:

[0049] Edge detection: used to identify tissue boundaries and ensure clear edge contours.

[0050] Texture analysis: used to assess the homogeneity and structural characteristics within tissues.

[0051] Morphological processing: It is used to remove noise and enhance important details in the image.

[0052] Statistical analysis: used to quantify characteristic parameters, such as mean gray value, standard deviation, etc.

[0053] ③ Tissue content calculation: Select the area where the tissue of interest is located in the single-frequency thermoacoustic image, extract the characteristic parameters in the area, and calculate the fat content. Calculate the fat tissue content based on the characteristic parameters, and compare and verify the result with the actual fat content.

[0054] 9. Identification of lesions and tissue characteristics: Analyze relevant lesions based on the quantified content of different tissue components; at the same time, by comparing multi-frequency fusion images, highlight those tissue areas and corresponding pathological conditions that cause significant changes in microwave energy absorption with frequency changes.

[0055] The main advantage and innovative effect of the present invention lies in the use of multi-frequency microwave thermoacoustic imaging technology, based on the difference in the absorption of microwave energy of different frequencies by biological tissues, by carefully selecting multiple frequency ranges with significant absorption differences as excitation sources. Specifically, for tissue types that show weak microwave absorption under single-frequency excitation but have large changes in absorption coefficient within a specific frequency range, we design the system to use the boundary frequency points of these frequency intervals as the core parameters of multi-frequency excitation.

[0056] In the experimental operation, the system uses a series of pulsed microwave signals of different frequencies to stimulate the biological tissue under test in turn. The thermoacoustic signal generated by the tissue at each frequency is detected by the ultrasonic transducer, and after being enhanced by the amplifier, it is transmitted to the data acquisition card and imaged in real time on the computer. For each selected frequency, the corresponding single-frequency thermoacoustic image is reconstructed and fused into a multi-frequency thermoacoustic image.

[0057] Among them, each set of single-frequency images is used to deeply analyze the biological tissues and related lesions that have strong microwave energy absorption at this frequency; by integrating all single-frequency thermoacoustic images through a multi-frequency fusion algorithm, a high-contrast multi-frequency thermoacoustic image can be obtained. This image can significantly highlight those tissue structures and their potential lesions that have weak microwave absorption at a single frequency but have large fluctuations in absorption characteristics with frequency changes.

[0058] In addition, the present invention also realizes the function of calculating and quantitatively evaluating the content of biological tissue components. By analyzing multiple single-frequency thermoacoustic imaging data, the content of different tissue components (such as fat, muscle, blood vessels, etc.) in the sample can be accurately determined, enhancing the detection ability of the internal structure and functional state of complex biological tissues.

[0059] In summary, compared with the traditional single-frequency thermoacoustic imaging technology, the multi-frequency microwave thermoacoustic imaging system proposed in the present invention shows excellent performance in revealing the structural characteristics of tissues with weak microwave absorption but strong frequency dependence and realizing the quantification of tissue components, which greatly improves the imaging quality and lesion recognition rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of a dual-frequency microwave thermoacoustic imaging system according to a preferred embodiment of the present invention.

[0061] Figure 2 It is a schematic diagram of the overall process of the tissue identification method in an embodiment of the present invention.

[0062] Figure 3 This is the step for executing multi-frequency microwave thermoacoustic imaging in an embodiment of the present invention.

[0063] Figure 4 It is a flow chart for calculating the tissue contents of different components provided in an embodiment of the present invention.

[0064] Figure 5 This is a real picture of pork belly.

[0065] Figure 6 This is an example diagram of tissue content calculation. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example

[0068] The present invention aims to solve the limitations of existing microwave thermoacoustic imaging technology and proposes a multi-frequency microwave thermoacoustic imaging method to realize the calculation of different tissue contents. The core of this method and system is the ability to generate high-resolution images of multiple frequency information. Its uniqueness lies in the ability to effectively highlight those tissue structures that have weak energy absorption at a single frequency but show significant absorption characteristics as the frequency changes, and realize the calculation of different tissue contents.

[0069] As a preferred embodiment, this embodiment provides a multi-frequency microwave thermoacoustic imaging system, which includes:

[0070] An image acquisition system is configured to acquire thermoacoustic signals of a sample under at least two frequency excitations, reconstruct a single-frequency thermoacoustic image through an imaging algorithm, and obtain a multi-frequency thermoacoustic image through a fusion algorithm;

[0071] The image analysis module is configured to calculate the content of different tissues in the sample through tissue thermoacoustic signals under at least two frequency excitations.

[0072] Image acquisition system such as Figure 1 As shown, it includes a microwave excitation module, an ultrasonic coupling module, a data acquisition module, and a real-time imaging module, specifically, a multi-frequency microwave source, an antenna, an ultrasonic detector, an amplifier and a multiplexer, a data acquisition card, and a computer. The multi-frequency microwave source is used to generate pulsed microwaves of different frequencies, and the microwaves are radiated to the biological tissue under test through the antenna. After the biological tissue absorbs the microwave energy, it shrinks and expands to generate thermoacoustic signals. The thermoacoustic signals under the excitation of microwaves of different frequencies are successively detected by the ultrasonic transducer, and are amplified by the amplifier and transmitted to the multi-channel acquisition card, thereby obtaining multiple groups of thermoacoustic data under different frequency excitations, which can be imaged in real time on the computer, and the entire process is controlled by the computer. After data acquisition, the single-frequency thermoacoustic image is reconstructed using algorithms such as back-projection algorithm, time reversal algorithm, and compressed sensing algorithm according to the relationship between time domain information and spatial position, and then the single-frequency thermoacoustic data or image is fused using a fusion algorithm to obtain a multi-frequency thermoacoustic image. The frequency in the multi-frequency microwave induction refers to the microwave center frequency. The single-frequency thermoacoustic image is used to analyze biological tissues with strong single-frequency microwave energy absorption and detect corresponding lesions and calculations of different tissue contents, while the multi-frequency fusion image is used to highlight tissues with weak single-frequency microwave energy absorption but large changes in microwave energy absorption with frequency changes and detect corresponding lesions.

[0073] The present invention is based on the bioelectromagnetic law that the microwave absorption coefficient between biological tissues changes with frequency. For tissues with weak microwave absorption under single-frequency microwave excitation, a frequency range in which the microwave absorption coefficient of the tissue changes greatly while the microwave absorption coefficient of the surrounding tissue changes less is selected, and then the boundary frequency point of this frequency range is selected as the excitation frequency point for multi-frequency microwave thermoacoustic imaging.

[0074] Multi-frequency microwaves are pulsed microwaves of multiple different frequencies. During the experiment, the biological tissue under test is excited through the antenna in turn, and the generated thermoacoustic signals are detected by the ultrasonic detector in turn. The detected thermoacoustic signals are amplified by the amplifier and collected by the data acquisition card in turn. Image reconstruction of the thermoacoustic signals detected under microwave excitation of different frequencies can obtain multiple groups of thermoacoustic images under microwave excitation of different frequencies, and the multiple groups of thermoacoustic images are fused to obtain a group of multi-frequency thermoacoustic images.

[0075] Multi-frequency microwaves can be generated by the same variable-frequency microwave source or by multiple single-frequency microwave sources, and the pulse width and peak power of the generated microwaves must be kept consistent. The antenna can be a non-frequency-variable antenna covering a dual-frequency range or for multiple required radiation frequency points, or a wide-band antenna covering a multi-frequency range, and the field distribution of multiple frequency points used for excitation imaging must be kept consistent. When switching multi-frequency microwave excitation, the antenna and the biological tissue under test remain unchanged, that is, other experimental conditions except the excitation microwave frequency remain unchanged.

[0076] The entire multi-frequency microwave thermoacoustic imaging system includes a microwave excitation module, an ultrasonic coupling module, a data acquisition module and a real-time imaging module. The microwave excitation module includes a multi-frequency microwave excitation source and an antenna. The ultrasonic coupling module includes an ultrasonic transducer, a coupling medium and biological tissue. Among them, the coupling medium acts as a microwave transmission medium from the radiation antenna to the tissue and an ultrasonic transmission medium from the tissue to the ultrasonic transducer, and simultaneously affects the transmission efficiency of microwave energy and ultrasonic waves. The data acquisition module includes a data acquisition card, an amplifier and a multiplexer. The real-time imaging module includes a computer, on which the thermoacoustic image can be viewed in real time and the thermoacoustic signal data can be stored.

[0077] Single-frequency thermoacoustic image reconstruction can use traditional microwave thermoacoustic imaging algorithms such as back-projection algorithm, time reversal algorithm and compressed sensing algorithm, and it is only necessary to ensure that the same algorithm is used to reconstruct the corresponding single-frequency thermoacoustic data. The multi-frequency fusion algorithm is based on single-frequency thermoacoustic imaging, mainly involving differential fusion, and the weight coefficient of differential fusion is comprehensively determined based on the theoretical microwave absorption coefficient and the imaging smoothness or focusing effect of the tissue extracted and reconstructed to highlight. The multi-frequency fusion algorithm can be performed on the basis of the reconstructed single-frequency thermoacoustic image, or it can directly process the single-frequency thermoacoustic data.

[0078] The multi-frequency microwave source can be composed of independent microwave sources that can output the same pulse and the same power but different frequencies, or it can be a microwave source with variable frequency but ensuring that other pulse microwave parameters remain unchanged, among which a variable frequency microwave source is preferred.

[0079] The antenna can be a wideband antenna covering multiple frequency ranges, but the frequency point should not be at the boundary frequency point of the wideband antenna as much as possible. It can also be a non-frequency-variable antenna targeting multiple required radiation frequency points or covering multiple frequency ranges, among which the non-frequency-variable antenna is preferred.

[0080] Except for the frequency, other parameters of the pulse microwave remain unchanged. The pulse microwave excites the measured tissue through a coaxial line of the same length with consistent loss parameters and an antenna with consistent field distribution. The information reflected by the multi-frequency image relies more on the change of the tissue's own intrinsic microwave absorption coefficient and reduces the impact of changes in external field distribution.

[0081] The ultrasonic coupling module detects the thermoacoustic signals under different frequency excitations, and the data acquisition module uses multiplexing technology to ensure real-time imaging, thereby avoiding jitter interference during in vivo imaging.

[0082] This system can perform in vivo, ex vivo and living imaging. When performing multi-frequency imaging to collect thermoacoustic signals under different microwave frequency excitations, other experimental conditions except the frequency are kept as unchanged as possible.

[0083] The multi-frequency fusion algorithm can be performed on the basis of the reconstructed single-frequency thermoacoustic image, or can directly process multiple sets of single-frequency thermoacoustic data, and the thermoacoustic image finally obtained includes multiple sets of single-frequency thermoacoustic images and one set of multi-frequency thermoacoustic image. Among them, the multi-frequency fusion is preferably performed on the basis of the reconstructed multiple sets of single-frequency thermoacoustic images.

[0084] The overall process of the method of the present invention is as follows Figure 2 As shown, specifically:

[0085] Multi-frequency microwave thermoacoustic imaging methods such as Figure 3 As shown, it includes the following steps:

[0086] Step 1: For biological tissues that have weak microwave absorption under single-frequency excitation and are expected to be highlighted by multi-frequency thermoacoustic imaging, first determine the frequency range in which their theoretical microwave absorption coefficient changes significantly while the microwave absorption coefficient of surrounding tissues is relatively stable, and select a representative boundary frequency point within this range as one of the multiple excitation frequency points for multi-frequency microwave thermoacoustic imaging.

[0087] Step 2: Start and initialize all relevant equipment and set system parameters to meet experimental requirements;

[0088] Step 3, firmly fixing the biological tissue to be tested in an appropriate position;

[0089] Step 4: Use a computer program to control the multi-frequency microwave source to generate a pulsed microwave pulse of the first center frequency, and start the data acquisition card at the same time: the microwave pulse of the specific frequency is accurately radiated to the biological tissue to be tested via the antenna, and the tissue generates an ultrasonic signal (i.e., a thermoacoustic signal) based on the thermoacoustic effect. The thermoacoustic signal is captured by the ultrasound detector and converted into an electrical signal, which is then collected by the data acquisition card and stored in the computer for corresponding single-frequency thermoacoustic image reconstruction processing, thereby completing the construction of thermoacoustic imaging at the first frequency;

[0090] Step 5: Keep the experimental environment and settings unchanged to ensure the consistency of all parameters;

[0091] Step 6: The multi-frequency microwave source is again instructed by the computer program to generate another pulse microwave of the center frequency, and the data acquisition card is started synchronously: another microwave pulse of a specific frequency is also transmitted to the biological tissue to be tested through the antenna, and the ultrasonic signal generated by the thermoacoustic effect is detected by the ultrasonic detector and converted into an electrical signal. These data are again acquired by the acquisition card and stored in the computer, and then the single-frequency thermoacoustic image reconstruction operation of the thermoacoustic signal at this frequency is implemented;

[0092] Step 7: Use a multi-frequency image fusion algorithm to post-process and fuse the single-frequency thermoacoustic image of the biological tissue to be tested, or directly perform fusion operation processing on the single-frequency thermoacoustic data at different frequencies stored in the computer;

[0093] Step 8. The final image output is a single-frequency thermoacoustic image of each frequency, which is used to analyze and identify biological tissues and their pathological characteristics that have strong microwave energy absorption under a single frequency; while the multi-frequency fusion image focuses on displaying the tissue structure and its corresponding pathological information that have weak microwave absorption under single-frequency excitation but have large differences in absorption characteristics with frequency changes, thereby achieving all-round and high-precision visualization of biological tissue components and pathological detection.

[0094] like Figure 3 As shown, this method of calculating the content of different tissues by multi-frequency microwave thermoacoustic imaging is as an example. Assume that the frequency is dual-frequency, the two tissues to be measured (i.e., tissue F and tissue M described in the Summary of the Invention) are fat (equivalent to tissue F) and muscle (equivalent to tissue M), respectively. The thermoacoustic pressure generated by the heat source H(r, t) follows the following equation:

[0095]

[0096] in represents the sound pressure distribution, c represents the speed of light, β represents the volume expansion coefficient, c p is the specific heat capacity.

[0097] The equation for the thermoacoustic pressure leads to the forward problem:

[0098]

[0099] Where r-r' is the distance between the source point and the field point, and t represents time.

[0100] The heat source is modeled as the product of two factors: the spatial distribution of energy absorption A(r), which is characteristic of the object being imaged; and the temporal illumination function I(t). Because the ultrasound transducer array has a finite bandwidth, the recorded thermoacoustic measurements is the convolution of the induced pressure and the impulse response h(t) of the ultrasound transducer array.

[0101]

[0102] When the object of interest is heated with a pulse of microwave radiation, the power deposition per unit volume is expressed as

[0103]

[0104] Where ω represents the microwave angular frequency, ε0 ​​and ε r ″ represent the dielectric constant and relative dielectric constant in vacuum, respectively. Represents the microwave electric field.

[0105] The electrical conductivity of the tissue is σ=ωε0ε″ r , so Therefore, the thermoacoustic signal intensity TAS obtained from the tissue is the product of the deposited energy of the tissue and the Grüneisen parameter Γ.

[0106]

[0107] Based on the thermoacoustic signals collected by dual-frequency microwave excitation with strong absorption coefficients for liver fat (fat) and non-fat tissue (set as muscle), the relative concentrations of fat and muscle can be obtained:

[0108]

[0109] In the above formula, and C Muscle Respectively represent fat and muscle content, σ Fat (f) and σ Muscle (f) is the microwave absorption coefficient of Fat and Muscle, and TAS is the measured thermoacoustic signal intensity. Based on the data obtained from the dielectric property measurement, if multi-frequency excitation is used for imaging, according to the formula:

[0110]

[0111] f n Represents the nth frequency. If dual-frequency excitation is used, the formula for calculating the proportion of fat tissue in the sample is:

[0112]

[0113] When a microwave source with center frequencies of f1 and f2 (e.g., f1 = 1 GHz, f2 = 3 GHz) is selected to excite tissue, if only the effects of fat and muscle are considered, the absorption intensity at the two frequencies can be expressed as

[0114]

[0115] Where CFAT , C MUSCLE Represent the content of Fat and Muscle respectively, σ is the conductivity of the tissue, is the intensity of the electric field, and Γ is the Gruneisen parameter of the tissue.

[0116] Assuming that the intensity of the thermoacoustic signal is proportional to the absorption coefficient of the tissue, the simplified calculation can be obtained:

[0117]

[0118] Where σ m1 and σ m2 is the conductivity of muscle tissue at frequencies f1 and f2, σ f1 and σ f2 is the conductivity of fat at frequencies f1 and f2, E1 and E2 are the electric field strengths at frequencies f1 and f2, Γ f and G m are the Gruneisen parameters of fat and muscle, respectively. A1 and A2 are the thermoacoustic signal intensity information at frequencies f1 and f2, respectively.

[0119] like Figure 4 As shown, the method for calculating the content of different tissues by multi-frequency microwave thermoacoustic imaging includes the following steps:

[0120] Step A: When collecting the thermoacoustic signal, a saline tube reference that is insensitive to changes in the microwave center frequency is placed next to the tissue being tested, and used as a standardized tool to measure and calculate the electric field intensity at each excitation frequency.

[0121] Step B: Using an imaging algorithm to reconstruct the collected thermoacoustic signals to obtain multiple single-frequency thermoacoustic images, which correspond to tissue response characteristics under different microwave frequencies.

[0122] Step C: Based on the obtained single-frequency thermoacoustic image, select the location of the brine pipe in the image and calculate the normalized electric field intensity, assuming that the electric field at each frequency is uniformly distributed.

[0123] Step D: In each single-frequency thermoacoustic image, the region where the tissue of interest is located is selected, and feature extraction is performed. Based on the extracted feature parameters, the content of fat tissue in the target region is calculated, and the result is compared and verified with the actual fat content.

[0124] Step E: Integrate the data on the content of different tissue components obtained from the above quantitative analysis, and combine it with the rich information provided by the multi-frequency fusion image to deeply analyze the tissue structure and lesion conditions, thereby providing a more comprehensive and accurate imaging basis for clinical diagnosis, treatment decision-making and scientific research exploration.

[0125] like Figure 6 As shown, Figure 5 Dual-frequency microwave thermoacoustic imaging and content calculation of pork belly. Figure 6 (a)(c)(e)(g) are the thermoacoustic images and calculation process of fat and muscle at 1 GHz microwave frequency. Figure 6 (b)(d)(f)(h) are the thermoacoustic images and the calculation process of fat and muscle at 3GHz microwave frequency. The program calculated the fat content to be 55.676% and the muscle content to be 44.324%. In reality, fat: (1÷1.8)×100%=55.56%, muscle: (0.8÷1.8)×100%=44.44%. The program calculated results are not much different from the actual calculated results, which is in line with expectations. This phenomenon strongly confirms the feasibility of multi-frequency microwave thermoacoustic imaging technology in quantifying the content of different biological tissues. Since this technology can capture and analyze the differences in the absorption of microwave energy by tissues at multiple frequencies, the acquired thermoacoustic image data can be accurately simulated and measured by processing the acquired thermoacoustic image data with a carefully designed algorithm. When the program calculation results maintain a high degree of consistency with the actual experimental values, it means that the multi-frequency microwave thermoacoustic imaging method we use has good stability and accuracy, which is sufficient to be applied in clinical and scientific research fields for quantitative analysis and lesion detection of different tissue types such as fat, muscle, and blood vessels, thereby greatly promoting the progress and development of medical imaging technology and biological tissue research.

[0126] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0127] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0128] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A multi-frequency thermoacoustic imaging system, characterized in that: include: An image acquisition system is configured to acquire thermoacoustic signals of a sample under at least two frequency excitations, reconstruct a single-frequency thermoacoustic image through an imaging algorithm, and obtain a multi-frequency thermoacoustic image through a fusion algorithm; The image analysis module is configured to calculate the content of different tissues in the sample through tissue thermoacoustic signals under at least two frequency excitations.

2. The multi-frequency thermoacoustic imaging system according to claim 1, characterized in that: The tissue is selected from fat, muscle or blood vessel.

3. The multi-frequency thermoacoustic imaging system according to claim 1, characterized in that: The multi-frequency thermoacoustic image is a dual-frequency thermoacoustic image, and the calculation formula of tissue content is: Among them, F and M refer to two different tissues in the sample, TA_F is the proportion of tissue F in the sample, and C F and C M are the contents of the two tissues, σ m1 and σ m2 is the conductivity of tissue M at frequencies f1 and f2, σ f1 and σ f2 is the conductivity of tissue F at frequencies f1 and f2, E1 and E2 are the electric field strengths at frequencies f1 and f2, Γ f and Γ m are the Gruneisen parameters of tissue F and tissue M respectively, and A1 and A2 are the thermoacoustic signal intensity information at frequencies f1 and f2 respectively.

4. The multi-frequency thermoacoustic imaging system according to claim 1, characterized in that: The at least two frequencies are at least two frequencies having different microwave center frequencies but the same other parameters, wherein the other parameters include pulse width, pulse power and pulse repetition frequency.

5. The multi-frequency thermoacoustic imaging system according to claim 1, characterized in that: The imaging algorithm is selected from a back-projection algorithm, a delay and superposition algorithm, a filtered back-projection algorithm, a time reversal algorithm or compressed sensing.

6. The multi-frequency thermoacoustic imaging system according to claim 1, characterized in that: The fusion algorithm obtains a multi-frequency thermoacoustic image through single-frequency thermoacoustic data or images, specifically including: fusing multiple groups of single-frequency thermoacoustic images or directly fusing multiple groups of single-frequency thermoacoustic data, the fusion adopts differential fusion, the initial fusion weight factor is determined by the theoretical microwave absorption coefficient, after the multi-frequency thermoacoustic image is obtained, the smoothness and focus intensity of the tissue area to be highlighted are judged, if the tissue is smooth and focused, the multi-frequency thermoacoustic image obtained after fusion is directly output and accompanied by several groups of single-frequency images, if the tissue is not smooth and not focused, the weight factor is adjusted until the tissue is smooth and focused, and then a group of multi-frequency thermoacoustic images is output and accompanied by multiple groups of single-frequency thermoacoustic images.

7. The multi-frequency thermoacoustic imaging system according to any one of claims 1 to 6, characterized in that: The image acquisition system includes: a multi-frequency microwave source, a coaxial line, an antenna, an ultrasonic transducer, an amplifier, a multi-channel data acquisition card and a computer. The multi-frequency microwave source is used to generate pulse microwaves of different frequencies, which are transmitted to the antenna via a coaxial cable, and the microwaves are radiated to the sample under test. The sample under test absorbs microwave energy to generate ultrasonic waves, i.e., thermoacoustic signals. The thermoacoustic signals under microwave excitation of different frequencies are detected by ultrasonic transducers respectively, and the detected signals are amplified by amplifiers. The amplified signals are collected by multi-channel data acquisition cards, thereby obtaining thermoacoustic data under excitation of different frequencies. The entire process is controlled by a computer, and a single-frequency thermoacoustic image is reconstructed using an imaging algorithm according to the relationship between time domain information and spatial position. Then, a multi-frequency thermoacoustic image is obtained by using a fusion algorithm for the single-frequency thermoacoustic data or image. The frequency in the multi-frequency microwave thermoacoustic imaging refers to the microwave center frequency. The single-frequency thermoacoustic image is used to analyze biological tissues with strong single-frequency microwave energy absorption, and the area of ​​interest in the single-frequency image is selected to calculate the content of different tissue components. The multi-frequency thermoacoustic image is used to highlight tissues with weak single-frequency microwave energy absorption and large microwave energy absorption changes with frequency changes, and to detect corresponding lesions.

8. The multi-frequency thermoacoustic imaging system according to claim 7, characterized in that: The pulse microwaves of different frequencies generated by the multi-frequency microwave source are generated by the same variable-frequency microwave source, or by multiple single-frequency microwave sources; the pulse microwaves of different frequencies need to maintain the same pulse width and peak power; or the multi-frequency microwave source has a power monitoring and real-time feedback function; And / or, the antenna is a non-frequency-variable antenna covering multiple frequency ranges or for multiple required radiation frequency points; or, the antenna is a wide-band antenna covering multiple frequency ranges; the antenna needs to ensure that the field distribution of multiple frequency points used for excitation imaging is consistent; the antenna with consistent field distribution excites the sample under test; And / or, the ultrasonic transducer, amplifier, and multi-channel data acquisition card constitute a thermoacoustic signal acquisition module, and when the multi-frequency microwave excitation is switched, the ultrasonic transducer, amplifier, multi-channel data acquisition card, antenna, and biological tissue under test remain stationary; And / or, the multi-frequency microwave source and the antenna constitute a microwave excitation module, wherein the multi-frequency microwave source is composed of multiple independent microwave sources that can output the same pulse width, the same repetition frequency, the same power and different frequencies, or the multi-frequency microwave source is a microwave source with a variable frequency but ensuring that other pulse microwave parameters remain unchanged, and the other pulse microwave parameters include pulse width, pulse power and pulse repetition frequency.

9. A multi-frequency microwave thermoacoustic imaging method based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: When the multi-frequency is dual-frequency, for biological tissues with weak microwave absorption under single-frequency microwave excitation, a frequency range in which the theoretical microwave absorption coefficient changes greatly but the microwave absorption coefficient of surrounding tissues changes less is selected, and two boundary frequency points of this frequency range are selected as two excitation frequency points for dual-frequency microwave thermoacoustic imaging; Step 2: Turn on the device and set parameters for initialization; Step 3: Place the sample to be tested into the coupling liquid; Step 4: Use a computer to trigger the multi-frequency microwave source to generate a pulse microwave with a center frequency, and trigger the data acquisition card to start working: a pulse microwave with a center frequency is radiated to the biological tissue to be tested through the antenna, and the tissue generates a thermoacoustic signal based on the thermoacoustic effect. The thermoacoustic signal is detected by the ultrasound detector, converted into an electrical signal by the ultrasound detector, and collected by the data acquisition card, and then stored in the computer and processed by single-frequency thermoacoustic image reconstruction; Step 5: Keep all experimental conditions unchanged; Step 6: Use a computer to trigger the multi-frequency microwave source to generate another pulse microwave with a center frequency, and trigger the data acquisition card to start working: another pulse microwave pulse with a center frequency is radiated to the biological tissue to be tested through the antenna, and the tissue generates ultrasonic waves, i.e., thermoacoustic signals, based on the thermoacoustic effect. The thermoacoustic signals are detected by the ultrasonic detector, converted into electrical signals by the ultrasonic detector, and collected by the data acquisition card, and then stored in the computer and processed by single-frequency thermoacoustic image reconstruction, that is, the thermoacoustic imaging of this single-frequency excitation is completed; according to actual needs, more than two frequency microwaves can be generated to excite thermoacoustic signals; Step 7: Use the fusion algorithm to post-process the completed single-frequency thermoacoustic image, or directly process the single-frequency thermoacoustic data stored in the computer; Step 8: Export images, where the single-frequency thermoacoustic image is used to analyze biological tissues with strong single-frequency microwave energy absorption and detect corresponding lesions, while the multi-frequency thermoacoustic image is used to highlight tissues with weak single-frequency microwave energy absorption but large changes in microwave energy absorption with frequency changes and detect corresponding lesions.

10. A method for calculating tissue content based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step A: collecting the electric field intensity of a reference object that is insensitive to the change of microwave center frequency at each excitation frequency as the normalized base of the multi-frequency microwave thermoacoustic signal; Step B: acquiring multiple single-frequency thermoacoustic images; Step C: calculating the normalized electric field intensity of the single-frequency thermoacoustic image; Step D: Extract tissue features from the single-frequency thermoacoustic image to obtain the area where the tissue of interest is located. Step E: Calculate the content of different tissues based on the selected area where the tissue of interest is located.

Images