A human variable cross-section bone thickness detection method and system based on ultrasonic guided waves

Abstract

The application discloses a human variable cross-section bone thickness detection method and system based on ultrasonic guided waves, the system derives a human cortical bone dispersion curve through an anisotropic model, obtains an inverted cortical bone thickness distribution rule by using the dispersion curve, and further obtains a cortical bone two-dimensional thickness distribution image, and the system has no radiation hazard to patients, is low in cost, high in real-time performance, wide in application range, and convenient and fast in obtaining bone thickness measurement results, and takes into account the uneven human bone thickness distribution, and has important value in practice.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of ultrasonic technology and medical signal processing, and relates to a human variable cross-section bone thickness detection method and system based on ultrasonic guided waves. BACKGROUND

[0002] At present, the main means for diagnosing osteoporosis is based on radiation technology. DXA is the "gold standard" for diagnosing bone conditions today, which can provide very accurate data on bone mineral density (BMD) and bone shape, but it cannot give a comprehensive information on bone conditions, such as bone mechanical properties and bone microstructure. The bone quality measurement method based on ultrasonic guided waves has the advantages of no radiation, low cost, and sensitivity to bone microstructure. However, for the current ultrasonic guided wave bone quality detection research, many ultrasonic guided wave detection methods ignore the anisotropy of cortical bone material, resulting in errors in the measurement results. At the same time, the current detection method for bone thickness measurement is only limited to the average thickness, and the thickness of long bone is assumed to be uniform, so the result cannot reflect the distribution change rule of bone thickness with space. SUMMARY

[0003] In order to overcome the above problems, the present inventors have made intensive research and developed a human variable cross-section bone thickness detection method and system based on ultrasonic guided waves. The system derives the human cortical bone dispersion curve through an anisotropic model, obtains the inverted cortical bone thickness distribution rule by using the dispersion curve, and further obtains the two-dimensional thickness distribution image of human cortical bone. The system has no radiation hazard to patients, is low in cost, high in real-time performance, wide in application, and convenient and fast in obtaining bone thickness measurement results, and takes into account the uneven distribution of human bone thickness, and has important value in practice. The system provided by the application obtains the thickness change distribution of the cortical bone layer of long bones such as tibia by in vitro detection, is used for portable in vivo detection of bone quality, provides a basis for judging osteoporosis and other orthopedic diseases, and thus the application is completed.

[0004] Specifically, the application aims to provide the following aspects:

[0005] In a first aspect, a human variable cross-section bone thickness detection method based on ultrasonic guided waves is provided, which comprises:

[0006] S100, obtaining an ultrasonic guided wave signal acting on human cortical bone;

[0007] S200, preprocessing the ultrasonic guided wave signal to separate out a direct A0 mode signal and a direct S0 mode signal;

[0008] S300, convert the time-space domain signals of the direct A0 mode signal and the S0 mode signal into wave number space domain signals to obtain a spatial wave number distribution of the A0 mode and a spatial wave number distribution of the S0 mode;

[0009] S400, obtain a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode;

[0010] S500, obtain an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function;

[0011] S600, substitute the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

[0012] In S200, the preprocessing includes: performing a band-pass filtering process on the obtained ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal.

[0013] The S200 includes the following steps:

[0014] S201, perform a band-pass filtering process on the obtained ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal;

[0015] S202, obtain the propagation time of the direct A0 mode signal and the propagation time of the direct S0 mode signal according to the propagation speed of the ultrasonic guided wave signal in the human cortical bone, and further intercept the filtered ultrasonic guided wave signal to obtain the direct A0 mode signal and the S0 mode signal.

[0016] In S300, the time-space domain signals of the direct A0 mode signal and the S0 mode signal are subjected to fast Fourier transform to obtain frequency domain signals arranged according to spatial positions; then, the frequency of the signal is fixed as the excitation signal frequency, a short spatial Fourier transform is used in the spatial domain, and a spatial window function is used to shift the multi-point ultrasonic guided wave signal to obtain the spatial wave number distribution of the A0 mode at different positions and the spatial wave number distribution of the S0 mode at different positions.

[0017] In S500, the wave number space ridge curve is substituted into the dispersion curve fitting function to obtain the inverted cortical bone thickness distribution rule.

[0018] In a second aspect, a human variable cross-section bone thickness detection system based on ultrasonic guided waves is provided, and the system includes:

[0019] A measurement module is configured to obtain an ultrasonic guided wave signal acting on human cortical bone;

[0020] The signal preprocessing module is configured to preprocess the ultrasonic guided wave signal to separate the direct A0 mode signal and the direct S0 mode signal.

[0021] The wave number domain conversion module is configured to convert the time-space domain signals of the direct A0 mode signal and the S0 mode signal into wave number space domain signals to obtain the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode.

[0022] The time-frequency ridge extraction module is configured to obtain the wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode.

[0023] The thickness inversion module is configured to obtain the inverted cortical bone thickness distribution rule according to the wave number space ridge curve and the dispersion curve fitting function.

[0024] The two-dimensional thickness imaging module is configured to substitute the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a two-dimensional cortical bone thickness distribution image.

[0025] The signal preprocessing module includes:

[0026] The filtering module is configured to perform band-pass filtering processing on the acquired ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal.

[0027] The separation module is configured to intercept the direct A0 mode signal and the S0 mode signal from the filtered ultrasonic guided wave signal according to the propagation time.

[0028] The wave number domain conversion module includes:

[0029] The frequency domain conversion module is configured to obtain the preprocessed ultrasonic guided wave signal, perform fast Fourier transform on the multi-array ultrasonic guided wave signal, and obtain the frequency domain signal arranged according to the spatial position.

[0030] The short space conversion module is configured to fix the frequency of the signal as the excitation frequency, use the short space Fourier transform in the spatial domain, and use the spatial window function to shift on the multi-point ultrasonic guided wave signal to obtain the wave number power spectrum at different positions.

[0031] The system further includes a processor configured to perform the steps of the human variable cross-section bone thickness detection based on the ultrasonic guided wave, including:

[0032] The ultrasonic guided wave signal is preprocessed to separate the direct A0 mode signal and the direct S0 mode signal.

[0033] The time-space domain signals of the direct A0 mode signal and the S0 mode signal are converted into wave number space domain signals to obtain the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode.

[0034] According to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode, a wave number space ridge curve is obtained by using a time-frequency ridge extraction algorithm;

[0035] According to the wave number space ridge curve and a dispersion curve fitting function, an inverted cortical bone thickness distribution rule is obtained.

[0036] Different angle wave number space ridge curves are substituted into corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

[0037] In a third aspect, an ultrasonic guided wave-based human variable cross-section bone thickness detection system is provided, and the system comprises a signal generator, a power amplifier, an excitation transducer, a receiving array transducer and a computer device.

[0038] The computer device comprises a memory and a processor, the memory is used to obtain a cortical bone two-dimensional thickness distribution image, and the processor is used to perform the following steps:

[0039] The ultrasonic guided wave signal is preprocessed to separate out a direct A0 mode signal and a direct S0 mode signal.

[0040] The time-space domain signals of the direct A0 mode signal and the S0 mode signal are converted into wave number space domain signals to obtain a spatial wave number distribution of the A0 mode and a spatial wave number distribution of the S0 mode.

[0041] According to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode, a wave number space ridge curve is obtained by using a time-frequency ridge extraction algorithm.

[0042] According to the wave number space ridge curve and a dispersion curve fitting function, an inverted cortical bone thickness distribution rule is obtained.

[0043] Different angle wave number space ridge curves are substituted into corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

[0044] The present application has the following beneficial effects:

[0045] (1) The human variable cross-section bone thickness detection system based on ultrasonic guided waves provided by the application contains a memory, the memory derives the cortical bone dispersion curve through a transverse isotropic model, the dispersion curve is used to obtain the inverted cortical bone thickness distribution rule, and then a two-dimensional cortical bone thickness distribution image is obtained, so that the bone thickness change condition is observed intuitively and accurately, and the system is non-radiation and low in cost.

[0046] (2) The human variable cross-section bone thickness detection method based on ultrasonic guided waves provided by the application is non-radiation to patients, low in cost, high in real-time performance, and wide in application.

[0047] (3) The human variable cross-section bone thickness detection method based on ultrasonic guided waves provided by the application considers the uneven distribution of human bone thickness, and realizes the two-dimensional imaging technology of ultrasonic guided wave detection of variable cross-section bone thickness. BRIEF DESCRIPTION OF DRAWINGS

[0048] Various other advantages and benefits of the present application will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included solely for purposes of illustrating the preferred embodiments and are not to be construed as a limitation of the present application. It should be readily understood that the drawings depicted are only some embodiments of the present application and that any other drawings similarly drawn can be derived from the included drawings without resorting to undue experimental effort.

[0049] In the drawings:

[0050] Figure 1 A flow chart of the human variable cross-section bone thickness detection method based on ultrasonic guided waves according to a preferred embodiment of the present application is shown;

[0051] Figure 2 A structure schematic diagram of the human variable cross-section bone thickness detection system based on ultrasonic guided waves according to a preferred embodiment of the present application is shown;

[0052] Figure 3 A flow chart of converting the time-space domain signal of ultrasonic guided waves into the wave number space domain signal to obtain the spatial wave number distribution of A0 mode and the spatial wave number distribution of S0 mode according to a preferred embodiment of the present application is shown;

[0053] Figure 4 A physical diagram of the concave cortical bone plate of the isolated bovine tibia in Example 1 is shown;

[0054] Figure 5 An effect diagram of the spatial wave number distribution of A0 mode at different positions and the spatial wave number distribution of S0 mode at different positions in Example 1 is shown;

[0055] Figure 6 A wave number space ridge curve diagram obtained in Example 1 is shown;

[0056] Figure 7 A three-dimensional graph of the angle, phase velocity, and product of frequency and thickness of Example 1 plotted in A0 mode is shown;

[0057] Figure 8 A wave number-frequency product dispersion curve graph of A0 mode and S0 mode in Example 1 at an angle of 0° between the ultrasonic wave propagation direction and the bone direction is shown;

[0058] Figure 9 An effect image of the cortical bone thickness distribution obtained by inversion in Example 1 is shown;

[0059] Figure 10 A schematic diagram of a measurement method of translating or rotating an ultrasonic probe at an acquisition end along a cortical bone surface is shown, which is a preferred embodiment of the present application;

[0060] Figure 11 A two-dimensional thickness distribution image of the cortical bone obtained in Example 1 is shown;

[0061] BRIEF DESCRIPTION OF THE DRAWINGS

[0062] 101 - signal generator;

[0063] 102 - power amplifier;

[0064] 103 - excitation transducer;

[0065] 104 - receiving array transducer;

[0066] 105 - computer device. DETAILED DESCRIPTION

[0067] The present application will be described in more detail below with reference to the accompanying drawings, in which: Figures 1 to 11 The specific embodiments of the present application will be described in more detail below. Although specific embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be accurately conveyed to those skilled in the art.

[0068] It should be noted that some terms are used in the description and claims to refer to certain components. Those skilled in the art can understand that the same component can be referred to by different terms. The description and claims of the present application do not distinguish components by differences in terms, but by differences in functions. As mentioned throughout the description and claims, "including" or "including" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment for implementing the present application, and the description is for the purpose of the general principles of the specification, not to limit the scope of the present application. The scope of protection of the present application is defined by the appended claims.

[0069] In the description of the present application, it should be noted that the terms "upper", "lower", "inner", "outer", "front", "back" and the like indicate the orientation or positional relationship based on the working state of the present application, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third", "fourth" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0070] In order to facilitate the understanding of the embodiments of the present application, the following will be further explained and described with specific examples combined with the accompanying drawings, and each drawing does not constitute a limitation on the embodiments of the present application.

[0071] In a first aspect, the present application provides a human variable cross-section bone thickness detection system based on ultrasonic guided waves, the system comprising a computer device 105, the computer device 105 comprising a memory and a processor, the memory being configured to obtain a cortical bone two-dimensional thickness distribution image, and the processor being configured to perform the steps of human variable cross-section bone thickness detection based on ultrasonic guided waves.

[0072] The current human variable cross-section bone thickness detection method ignores the anisotropy of cortical bone material. When using the dispersion curve of ultrasonic guided waves in cortical bone propagation, the isotropic assumption is used. This assumption is different from the actual material condition of the skeleton, resulting in errors in the measurement results of bone thickness. The system described in the present application derives the cortical bone dispersion curve by using an anisotropic model, i.e. a transverse isotropic model, and uses the dispersion curve to obtain the inverted cortical bone thickness distribution rule, and further obtains a human cortical bone two-dimensional thickness distribution image.

[0073] According to the present application, the memory comprises:

[0074] The measurement module is configured to obtain an ultrasonic guided wave signal acting on the human cortical bone;

[0075] a signal preprocessing module, configured to preprocess the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct S0 mode signal;

[0076] a wave number domain conversion module, configured to convert the time-space domain signal of the direct Ao mode signal and the direct S0 mode signal into a wave number space domain signal to obtain the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode;

[0077] a time-frequency ridge extraction module, configured to obtain a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode;

[0078] a thickness inversion module, configured to obtain an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function;

[0079] a two-dimensional thickness imaging module, configured to substitute the wave number space ridge curve at different angles into the corresponding dispersion curve fitting function to obtain a two-dimensional cortical bone thickness distribution image.

[0080] In the present application, the signal preprocessing module comprises:

[0081] a filtering module, configured to preprocess the acquired ultrasonic signal, specifically, to perform a band-pass filtering process on the acquired ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal;

[0082] a separation module, configured to intercept the direct Ao mode signal and the S0 mode signal from the filtered ultrasonic guided wave signal according to the propagation time.

[0083] In the present application, the wave number domain conversion module comprises:

[0084] a frequency domain transformation module, configured to acquire the preprocessed ultrasonic guided wave signal, to perform a fast Fourier transform on the multi-array ultrasonic guided wave signal to obtain a frequency domain signal arranged according to the spatial position;

[0085] a short space transformation module, configured to fix the frequency of the signal as the excitation frequency, to use a short space Fourier transform in the spatial domain, and to use a spatial window function to shift on the multi-point ultrasonic guided wave signal to obtain the wave number power spectrum at different positions.

[0086] In the present application, the ultrasonic guided wave is different from the conventional ultrasonic wave, the ultrasonic guided wave is formed by reflection between the upper and lower surfaces of each module and / or unit, and thus the ultrasonic guided wave is very sensitive to the thickness of each module and / or unit, compared with the conventional ultrasonic wave, the ultrasonic guided wave is more suitable for detecting the bone thickness, and the measurement accuracy is much higher than that of the echo detection method of the conventional ultrasonic wave.

[0087] In the present application, the system further comprises a signal generator 101, a power amplifier 102, an excitation transducer 103 and a receiving array transducer 104. The signal generator 101 is used to transmit an ultrasonic guided wave signal to a human long bone (or called bone plate) to be measured, and a sinusoidal Hanning window amplitude modulation pulse signal with a narrowband frequency domain is excited by the signal generator 101. The power amplifier 102 amplifies the energy of the pulse signal, and the excitation transducer 103 at the collection end excites ultrasonic guided waves in the long bone cortical bone. The array transducer 104 at the collection end receives the ultrasonic guided wave signal along the long bone shaft direction of the human body to be measured and transmits it to a computer device 105, as shown in Figure 2

[0088] In the present application, the receiving array transducer 104 is composed of N vibration sources, and the N is not less than 64. Among them, the number of vibration sources will affect the accuracy of the result. The data obtained by more than 64 vibration sources will cause a loss of half a spatial window at the beginning and end of the data in the later short spatial Fourier transform, and the result is more accurate and reliable. Too few vibration sources will lead to the limitation of the result.

[0089] Optionally, the excitation transducer 103 and the receiving array transducer 104 can be packaged into an integrated collection end ultrasonic probe, and the collection end ultrasonic probe is used to obtain the ultrasonic guided wave signal propagating axially in the long bone.

[0090] Further, the computer device 105 is connected with the signal generator 101, and is used to control the excitation and reception synchronization of the pulse signal.

[0091] In the present application, the processor is used to execute the following steps:

[0092] S200, pre-processing the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct S0 mode signal.

[0093] Specifically, the obtained ultrasonic guided wave signal is filtered according to the center frequency of the excitation narrowband signal; the filtered ultrasonic guided wave signal is cut according to the propagation time to obtain the Ao mode signal and the direct S0 mode signal, so as to exclude the interference of environmental noise and reflected signal on the original ultrasonic guided wave signal.

[0094] In a preferred embodiment, the S200 comprises the following steps:

[0095] S201, band-pass filtering the obtained ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal.

[0096] ​The ultrasonic guided wave signal is subjected to digital band-pass filtering processing, an input signal sequence of the digital band-pass filtering processing at time n is x(n), a response function of the digital band-pass filtering processing at time n is h(n), an output signal sequence of the digital band-pass filtering processing at time n is y(n), a Fourier transform of x(n) is X(e jw ), a Fourier transform of y(n) is Y(e jw ), and the output of the digital band-pass filtering processing is further represented as:

[0097]

[0098] Y(e jω )=X(e jω )H(e jω )

[0099] |Y(e jω )|=|X(e jω )||H(e jω )|

[0100] ∠Y(e jω )=∠X(e jω )+∠H(e jω )

[0101] In the formula, x(n) represents an input signal sequence of the digital band-pass filtering processing at time n, y(n) represents an output signal sequence of the digital band-pass filtering processing at time n, H(e jω ) represents a frequency response function of the digital band-pass filtering processing, x(m) represents an input signal at time m, and h(n-m)x(m) represents an influence generated by the input signal x(m) at time m at time n.

[0102] After the spectrum X(e jw ) of the input signal is subjected to digital band-pass filtering processing, it is converted into X(e jω )H(e jω ), and as long as a proper H(e jw ) is selected, X(e jω )H(e jω ) can meet the requirements, and the system function H(e jw ) of the digital band-pass filtering processing is set according to a band-pass range.

[0103] S202, according to a propagation speed of the ultrasonic guided wave signal in the human cortical bone, propagation time of the direct A0 mode signal and propagation time of the direct S0 mode signal are obtained, and the filtered ultrasonic guided wave signal is intercepted to obtain the direct A0 mode signal and the direct S0 mode signal.

[0104] S300, convert the time-space domain signals of the direct A0 mode signals and the S0 mode signals into wave number space domain signals to obtain the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode.

[0105] Specifically, fast Fourier transform is performed on the time-space domain signals of the N direct A0 mode signals and the N direct S0 mode signals to obtain frequency domain signals arranged according to spatial positions; then, the frequency of the signals is fixed as the frequency of the excitation signal, a short spatial Fourier transform is used in the spatial domain, and a spatial window function is used to shift the multi-point ultrasonic guided wave signals to obtain the wave number power spectrum of different positions, that is, the spatial wave number distribution of different positions of the A0 mode and the spatial wave number distribution of different positions of the S0 mode.

[0106] The number of the time-space domain signals of the ultrasonic guided waves is consistent with the number of vibration sources contained in the receiving array transducer, that is, one vibration source obtains one corresponding time-space domain signal of the direct A0 mode signal and one corresponding time-space domain signal of the direct S0 mode signal.

[0107] In an embodiment, as shown in Figure 3 S300 includes the following steps:

[0108] S301, perform fast Fourier transform on the time-space domain signals of the direct A0 mode signals and the S0 mode signals to obtain frequency domain signals arranged according to spatial positions, that is, a frequency space matrix;

[0109] S302, fix the frequency as the frequency of the excitation signal and take the spatial signal data at the frequency of the excitation signal;

[0110] S303, perform spatial fast Fourier transform on the spatial signal data at the frequency of the excitation signal with a sliding window to obtain the wave number power spectrum of different spatial positions.

[0111] The wave number power spectrum of different spatial positions is obtained by the following formula:

[0112]

[0113] In the formula, k represents the wave number, x represents the spatial position, y represents the initial position of the window function, t represents the time, i represents the imaginary symbol, ω represents the circular frequency, g(x, t) is the time signal recorded by the N vibration sources of the linear receiving array transducer, f0 represents the selected fixed frequency, h(y-α) represents the sliding window function in the x direction, α is the translation of the window function, and H(k, x, f0) represents the wave number power spectrum distribution result.

[0114] In an embodiment, taking the bovine tibia concave cortical bone plate as an example, the spatial wavenumber distribution of the A0 mode at different positions and the spatial wavenumber distribution of the S0 mode at different positions are as shown in FIG. 8A and FIG. 8B. Figure 5 As can be seen from FIG. 8A and FIG. 8B, the power spectrum amplitude of the spatial wavenumber distribution of the A0 mode at different positions is larger.

[0115] S400, obtaining a wavenumber space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wavenumber distribution of the A0 mode and the spatial wavenumber distribution of the S0 mode.

[0116] Specifically, the spatial wavenumber distribution of the A0 mode and the spatial wavenumber distribution of the S0 mode are compared, the spatial wavenumber distribution with a larger power amplitude is selected, the maximum value in the wavenumber power spectrum of the spatial wavenumber distribution with a larger power amplitude is extracted by using a time-frequency ridge extraction algorithm, and the maximum value is recorded as a wavenumber space ridge curve. The inventors have found that, generally, the power amplitude of the spatial wavenumber distribution of the A0 mode is larger, that is, the ultrasonic guided wave A0 mode is more sensitive to the thickness change of the cortical bone of long bone, and therefore, generally, the maximum value in the wavenumber power spectrum of the spatial wavenumber distribution of the A0 mode is extracted by using a time-frequency ridge extraction algorithm, so as to obtain a wavenumber space ridge curve.

[0117] More specifically, after the time-space domain signal of the ultrasonic guided wave is converted into a wavenumber space domain signal, some curves related to the characteristics of the wavenumber distribution will clearly or vaguely appear, the distribution patterns of these curves in the three-dimensional space are similar to “ridges”, and these curves are ridge lines or lines extending along the top of the ridge. Therefore, the line extending along the highest power point in the three-dimensional space in the wavenumber space curve is referred to as a wavenumber space ridge curve. The wavenumber space ridge curve is present on the wavenumber space surface, and is a line with different brightness and trend, which can represent the change rule of the wavenumber of the signal in the space.

[0118] In step S400, the energy of the ultrasonic guided wave signal in the wavenumber space domain can be expressed as:

[0119]

[0120] wherein, is the joint wavenumber space density of the signal, is a quadratic function of the signal, k represents the wavenumber, and x represents the spatial position. By fixing the spatial position and cyclically searching the recorded value at the maximum wavenumber energy density, the maximum wavenumber power of the spatial position can be obtained. By traversing all the spatial domain, the maximum wavenumber power of each spatial position can be obtained, and the wavenumber space energy ridge curve can be obtained.

[0121] S500, obtaining an inverted cortical bone thickness distribution rule according to the wavenumber space ridge curve and a dispersion curve fitting function.

[0122] According to the preferred embodiment, the S500 comprises the following steps:

[0123] S501, constructing a dispersion curve.

[0124] Specifically, according to the elastic dynamics theory, the sound control equation, the stress-strain equation and the strain-displacement equation in the anisotropic material derive the dispersion curve.

[0125] The sound control equation in the anisotropic material is expressed as:

[0126]

[0127] The strain-displacement equation is expressed as:

[0128] σ ij =c ijkl s kl

[0129] The strain-displacement equation is expressed as:

[0130]

[0131] Wherein, u i is the component of the displacement vector, s kl is the component of the strain tensor, σ ij is the component of the stress tensor, and the matrix c ijkl is the stiffness of the anisotropic material; i, j, k, l represent the tensor free index, which means that the index does not appear repeatedly in the same term and can be 1, 2, 3, such as σ ij represents any stress component; ρ represents the cortical bone density; is the partial differential symbol; x i represents one component of the wave vector.

[0132] In the present application, the guided wave propagation problem in the anisotropic material needs two sets of coordinate systems to describe, one is the material coordinate system to define the material parameters, and the other is the vibration coordinate system to describe the wave propagation and vibration. Without considering the propagation direction of the wave, in the material coordinate system of the cortical bone, according to the actual material characteristics of the cortical bone of the long bone, the material of the long bone cortical bone is simplified as a transversely isotropic material in the anisotropic material. For the transversely isotropic material, only 5 independent material variables are in the 9 material variables.

[0133] Wherein, the stiffness matrix c′ ijkl of the material coordinate system is composed as follows:

[0134]

[0135] Wherein, C 12 =C 13 ; C 23 =C 33 ; C55 = C 66 44 = (C 33 - C 23 ) / 2. C 11 , C 12 , C 13 denote the elastic constants that constitute the stiffness matrix.

[0136] When the orientation angle of the material coordinate system with respect to the vibration coordinate system is not 0, that is, when the propagation direction of the guided wave is not along the bone fiber direction, the stiffness matrix c' of the material coordinate system needs to be integrated into the vibration coordinate system, and the following coordinate transformation equation is used: ijkl

[0137] c ijkl = X(θ)c' ijkl X T (θ)

[0138] Wherein, X(θ) represents the coordinate transformation matrix;

[0139]

[0140] X T (θ) represents the transpose of X(θ); θ represents the angle between the material coordinate system and the vibration coordinate system; therefore, the stiffness matrix c of the vibration coordinate system can be obtained. ijkl

[0141] In the present application, the partial wave method is preferably used to solve the anisotropic material dispersion curve fitting function, which assumes that the partial wave propagating in an infinite medium has the following form:

[0142]

[0143] Wherein, u l represents the component of the partial wave sound field particle displacement vector; U l represents the amplitude of the plane simple harmonic wave; x1, x3 represent the components of the wave vector; t is time; α represents the ratio of the x3 component of the wave vector to the x1 component; i represents the imaginary symbol; k represents the wave number; c p represents the phase velocity of wave propagation.

[0144] Introducing it into the sound control equation, the following Christoffel equation is obtained:

[0145]

[0146] Wherein, the non-zero items in the matrix are as follows:

[0147] A 12 = C 16 ​​​+C 45 α 2

[0148] A 13 =(C 13 +C 55 )α,

[0149] A 23 =(C 36 +C 45 )α,

[0150] where A 11 , A 12 , etc. are elements of the Christoffel matrix; C 12 , C 22 , etc. are elastic constants that constitute the stiffness matrix; α is the ratio of the wave vector x3 component to the x1 component; c p is the phase velocity of wave propagation; and ρ represents the cortical bone density.

[0151] The non-trivial solution of [U1, U2, U3] can be obtained by setting the determinant of the A matrix to zero. The determinant of the A matrix is a 6th order polynomial of α given c p and c ijkl , and six solutions of α can be obtained. Substituting the six solutions of α into the Christoffel equation, six partial wave solutions, i.e. 6 basis solutions of [U1, U2, U3], can be obtained.

[0152] The acoustic field in the bone plate can be described as a linear combination of the six partial waves, which is expressed as:

[0153]

[0154]

[0155] where u l is the component of the particle displacement vector in the guided wave acoustic field; m represents the order of the six partial waves; σ I is the stress tensor component in the guided wave acoustic field; α m is the six solutions obtained from |A| = 0; U lm is the solution of [U1, U2, U3]; B m is the six unknown weighting coefficients that can be solved using boundary conditions; i is the imaginary unit; t is time; k is the wave number; x1 and x3 are the components of the wave vector; and ω is the circular frequency.

[0156] The specific expression of the coefficient matrix M Im can be calculated as:

[0157] M 1m= C 11 U 1m + C 13 U 3m α m + C 16 U 2m

[0158] M 2m = C 12 U 1m + C 23 U 3m α m + C 26 U 2m

[0159] M 3m = C 13 U 1m + C 33 U 3m α m + C 36 U 2m

[0160] M 4m = C 44 U 2m α m + C 45 (U 1m α m + U 3m )

[0161] M 5m = C 45 U 2m α m + C 55 (U 1m α m + U 3m )

[0162] M 1m = C 16 U 1m + C 35 U 3m α m + C 66 U 2m

[0163] Assuming that the upper and lower surfaces of the bone plate are free boundaries, and assuming that the thickness of the bone plate is h, the mathematical expression under the free stress boundary condition satisfies the boundary condition: when x3=0 and x3=h, σ 3l = 0, l = 1, 2, 3, we have:

[0164]

[0165] The A0 mode wave number-frequency thickness product dispersion curve at the angle of θ can be obtained by solving the above equation by numerical calculation, and the S0 mode wave number-frequency thickness product dispersion curve at the angle of θ can also be obtained, the dispersion curves at different angles are obtained by changing the angle of θ, that is, the dispersion curves of the transverse isotropic theory of cortical bone in A0 mode and S0 mode are obtained. In an embodiment, after obtaining the dispersion curve at one angle of θ, the dispersion curves at different angles of wave number and frequency thickness product are obtained by the wave number and phase velocity relationship, that is, the dispersion curves of the transverse isotropic theory of cortical bone in A0 mode and S0 mode are obtained, and the wave number and phase velocity relationship is represented as:

[0166]

[0167] In the formula, k is the wave number; f0 is the signal excitation frequency, which is 150 kHz in the embodiment; d is the thickness of the cortical bone to be measured; c p is the waveguide phase velocity; c p < f0, d > represents the dispersion relationship of frequency thickness product and phase velocity.

[0168] S502, fitting the dispersion curve to obtain a dispersion curve fitting function.

[0169] Specifically, the dispersion curves of the transverse isotropic theory of cortical bone in A0 mode and S0 mode are fitted as a target function to obtain a dispersion curve fitting function.

[0170] S503, substituting the wave number space ridge curve into the dispersion curve fitting function to obtain the inverted cortical bone thickness distribution.

[0171] S600, substituting the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a two-dimensional thickness distribution image of the cortical bone.

[0172] Specifically, S100 to S500 are repeated, the wave number space ridge curves obtained at different angles are substituted into the corresponding dispersion curve fitting functions, and a two-dimensional thickness distribution image of the cortical bone is obtained.

[0173] More specifically, in S600, the different angles refer to different angles between the waveguide measurement direction and the bone fiber direction (that is, the bone shaft axial direction).

[0174] In the present application, in S501, different angles refer to obtaining dispersion curves at different angles by changing the angle of θ, and in S600, different angles refer to measuring at different angles, that is, different angles between the waveguide measurement direction and the bone fiber direction (that is, the bone shaft axial direction). It can also be understood that the different angles in S600 should be brought into the dispersion curve fitting function of the corresponding angle in S501 for inversion.

[0175] According to the present application, the translation measurement method comprises moving the ultrasonic probe along the cortical bone surface in parallel; the angle measurement method comprises rotating the ultrasonic probe at the collection end with the center of the excitation transducer 103 as the center and the axial direction of the long bone diaphysis as the rotation axis, and the rotation angle θ is the included angle between the axial direction of the diaphysis and the direction of the ultrasonic probe at the collection end, as shown in Figure 10 .

[0176] In a second aspect, the present application provides a human variable cross-section bone thickness detection method based on ultrasonic guided waves, as shown in Figure 1 . The method comprises:

[0177] S100, obtaining an ultrasonic guided wave signal acting on human cortical bone;

[0178] S200, preprocessing the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct S0 mode signal;

[0179] S300, converting the time-space domain signals of the direct Ao mode signal and the S0 mode signal into wave number space domain signals to obtain the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode;

[0180] S400, obtaining a wave number space ridge curve using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode;

[0181] S500, obtaining an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function;

[0182] S600, substituting the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a two-dimensional cortical bone thickness distribution image.

[0183] Osteoporosis is a systemic bone disease that is prone to fracture, characterized by decreased bone mass and density, thinning or partial loss of trabeculae, and thinning of bone tissue cortex. Ultrasonic evaluation and diagnosis methods have the advantages of low cost, no ionizing radiation, simplicity, speed, portability, and are suitable for population census applications. It has received great attention in the evaluation of bone conditions and the diagnosis of osteoporosis.

[0184] The ultrasonic bone evaluation technology can be divided into two categories according to different measurement sites, trabecular bone measurement and cortical bone measurement. The measurement object of the trabecular bone is the calcaneus and the hip bone of the human body, etc.; the measurement object of the cortical bone is the long bone of the human body, including the tibia, the ulna and the radius, etc. Studies have shown that the cortical bone of the long bone is an excellent ultrasonic waveguide material. The ultrasonic guided wave in the long bone can propagate to the entire cortical layer, and is highly sensitive to the thickness of the cortical bone, which has a strong correlation with the bone density. By processing the collected ultrasonic guided wave signals in the time domain and the frequency domain, the thickness and the elastic velocity of the cortical bone and other parameters can be inversed to help diagnose osteoporosis.

[0185] In the prior art, due to the serious aliasing of the ultrasonic guided wave signal mode in the cortical bone, many scholars focus on the separation of the ultrasonic guided wave mode. For example, the signal is mapped from the time-distance domain to the frequency-wavenumber domain space by using two-dimensional Fourier transform; for example, the singular value decomposition method is used to extract the dispersion curve of the multi-ultrasonic guided wave mode; for example, the long bone ultrasonic guided wave signal is analyzed by using the blind source separation method, and the separation of the multi-mode signal is realized. Although these methods can ensure high frequency resolution and separate the multi-mode signal, they cannot reflect the size of the local wavenumber in the cortical bone, and thus the thickness change of the cortical bone cannot be obtained. In actual human bones, especially after the onset of osteoporosis, the thickness is uneven and changes, so there is a certain gap with the reality. The local wavenumber inversion method is used to obtain the local wavenumber of the measurement point in the present application, that is, the thickness of each measurement point can be inversed, and the change trend of the bone thickness can be directly reflected.

[0186] In addition, regarding the material properties of the cortical bone, it is known that the cortical bone of the long bone is not an isotropic material, but a transversely anisotropic material. However, the methods proposed by scholars are mostly based on the isotropic model, and the study on the material properties of the cortical bone is ignored. Simplifying the properties of the cortical bone material may cause certain systematic errors, especially for the high-order modes of the ultrasonic guided wave. The anisotropic model, i.e. the transverse isotropic model, is used to derive the dispersion curve of the cortical bone in the present application, and the bone thickness is inversed by using the dispersion curve, so that the change of the bone thickness is directly and accurately observed without radiation and low cost.

[0187] Embodiment

[0188] The present application will be further described below by specific examples, but these examples are only exemplary and do not constitute any limitation on the scope of protection of the present application.

[0189] Taking the measurement of the concave cortical bone plate of an isolated bovine tibia as an example, an isolated bovine tibia is selected, cut, defatted and bleached to prepare a cortical bone plate sample with a size of 100x30x3.58mm, and then a circular arc groove with a radius of 100mm and a depth of 2.26mm is cut on the sample, Figure 4Show the in vitro bovine tibial concave cortical bone plate physical map; S100, obtain the ultrasonic guided wave signal:

[0190] The signal generator 101 excites a sinusoidal Hanning window amplitude modulation pulse signal with a narrow-band frequency domain, and the excitation signal frequency is selected as 150 kHz according to the thickness of the measured cortical bone. After the signal energy is amplified by the power amplifier 102, the guided wave is excited in the cortical bone plate through the excitation transducer 103, and the guided wave signal is received through an array transducer 104 (64 array) arranged along the bone direction (for example, at an angle of 0 degrees).

[0191] S200, pretreat the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct So mode signal:

[0192] The ultrasonic guided wave signal is band-pass filtered on the computer processor, the filter range of the band-pass filter is set to 100-200 kHz according to the signal center frequency of 150 kHz, the guided wave signal is filtered through the digital band-pass filter, the propagation time of the two mode direct waves is calculated according to the propagation speed of the guided wave in the cortical bone (the Ao mode speed is about 1600 m / s, and the So mode speed is about 3000 m / s), and the filtered signal is intercepted to obtain the direct Ao and So mode signals.

[0193] S303, convert the time-space domain signals of the direct Ao mode signal and the So mode signal into wave number space domain signals to obtain the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the So mode, specifically:

[0194] S301, respectively, the direct Ao mode signal composed of 64 guided wave signals received by the 64 array and the So mode signal composed of 64 guided wave signals received by the 64 array are subjected to fast Fourier transform to obtain the frequency space matrix arranged according to the spatial position.

[0195] S302, obtain the frequency space matrix, fix the frequency equal to the excitation signal frequency 150 kHz, and take a column of spatial signal data at this frequency.

[0196] S303, the spatial signal data at the excitation frequency is subjected to spatial fast Fourier transform with a sliding window, and since the excitation signal is a Hanning window modulated sinusoidal pulse wave, the spatial window function is selected as a Hanning window, and the spatial window width is 32 data points, and each time the data points are slid by 1, the wave number power spectrum at different spatial positions is calculated.

[0197] Figure 5 Show the effect diagram of the spatial wave number distribution of the Ao mode at different positions and the spatial wave number distribution of the So mode at different positions in the embodiment;

[0198] S400, according to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode, a wave number space ridge curve is obtained by using a time-frequency ridge extraction algorithm:

[0199] According to relevant information, the guided wave A0 mode is more sensitive to the change of the cortical bone thickness of the long bone, so the A0 mode with larger power amplitude is selected for subsequent inversion. For the A0 mode, first fix the spatial position, and cyclically search the recorded value at the maximum wave number energy density, so as to obtain the maximum wave number power of the A0 mode at this spatial position. By traversing all spatial domains, the maximum wave number power of the A0 mode at each spatial position can be obtained, so as to obtain the wave number space energy ridge curve. Figure 6 A wave number space ridge curve obtained in embodiment 1 is shown.

[0200] S500, according to the wave number space ridge curve and the dispersion curve fitting function, an inverted cortical bone thickness distribution rule is obtained, specifically:

[0201] The input density of the bovine tibia is 1500 kg / m 3 , and the transverse isotropic 5 independent elastic constants are respectively C 11 = 24.76 Gpa; C 22 = C 33 = 17.5 Gpa; C 12 = C 13 = 10.69 Gpa; C 23 = 10.15 Gpa; C 55 = C 66 = 5.11 Gpa; C 44 = (C 33 -C 23 ) / 2 = 3.67 Gpa, and numerical calculation is used to solve to obtain the dispersion curve of the bone plate guided wave with the angle change. The dispersion curve with an included angle of 0-90° is calculated and reserved. Figure 7 A three-dimensional graph of the angle, phase velocity and frequency-thickness product relationship of the A0 mode is shown in embodiment 1; and according to the dispersion curve, the relationship between the wave number and the phase velocity is:

[0202]

[0203] In the formula, k is the wave number; f0 is the signal excitation frequency, which is 150 kHz in the embodiment; d is the thickness of the cortical bone to be measured; c p is the guided wave phase velocity; c p < f0, d > represents the dispersion relationship between the frequency-thickness product and the phase velocity, and the dispersion relationship between the phase velocity and the frequency-thickness product at different angles of each mode is substituted into the formula to obtain the dispersion curve between the wave number and the frequency-thickness product at different angles. Here, 0 degrees are taken as an example, Figure 8Fig. 1 shows the wave number-frequency thickness product dispersion curves of A0 mode and S0 mode at 0° in the ultrasonic wave propagation direction and the bone shaft direction in Example 1;

[0204] After that, the wave number-frequency thickness product dispersion curves of A0 mode and S0 mode at different angles are fitted as a target function (i.e. dispersion curve fitting function). The A0 mode wave number space ridge curve measured along the bone shaft direction (0°) is substituted into the 0° wave number-frequency thickness product dispersion target function for solving. Thus, the cortical bone thickness distribution result is obtained, Figure 9 Fig. 2 shows the regular effect diagram of the inverted cortical bone thickness distribution obtained in Example 1. In order to illustrate the superiority of using the transversely isotropic model, it is compared with the isotropic model. When the isotropic model calculates the dispersion curve, the input material parameters are as follows: the input density of bovine tibia is 1500 kg / m3, the two independent elastic constants of isotropy are C 11 22 33 12 13 23 44 55 66 11 12 , other processing steps are completely consistent, Figure 9 The smooth curve in the middle represents the true value, the circular curve represents the inversion result under the isotropic parameter, and the triangular curve represents the inversion result of transverse isotropy. It can be found that the thickness inversion error using the transverse isotropy assumption is reduced by 7.8% than the isotropic assumption, and the accuracy is higher.

[0205] S600, the wave number space ridge curves at different angles are substituted into the corresponding dispersion curve fitting function to obtain the two-dimensional thickness distribution image of the cortical bone:

[0206] First, the long bone shaft axial direction is found, and then the center of the excitation transducer is taken as the center of the circle, and the collection end ultrasonic probe is rotated. The rotation angle is the included angle between the shaft direction and the direction of the collection end ultrasonic probe. In this embodiment, the bone plate of the isolated bovine tibia is measured at -5°-5° variable angle, and the measurement is performed once every 1° change of the measurement angle. After 11 measurements, the above S100-S500 processing steps are repeated, the A0 mode wave number space ridge curves at different angles are substituted into the dispersion curve fitting function at the corresponding angle, and the results are plotted in the figure. Thus, the inverted two-dimensional thickness distribution image is obtained. Figure 11 ​​​​​​​​​​The cortical bone two-dimensional thickness distribution image obtained in Example 1 is shown. The result of inversion is the 30x12mm concave area in the center of the bovine tibia bone plate, and the color depth in the two-dimensional distribution image represents the measured thickness of the bovine tibia cortical bone plate. It can be found that the two-dimensional thickness image is a concave surface, and the positive and negative measurement errors of the concave groove of the bovine tibia bone plate are within ±0.35mm, which proves the correctness and intuitiveness of the method.

[0207] The above describes the present application in detail in combination with preferred embodiments and exemplary examples. However, it should be stated that these specific embodiments are only illustrative explanations of the present application, and do not constitute any limitation on the protection scope of the present application. Various improvements, equivalent replacements or modifications can be made to the technical content and embodiments of the present application without exceeding the spirit and protection scope of the present application, and these all fall within the protection scope of the present application. The protection scope of the present application is subject to the appended claims.

Claims

1. A method for detecting the thickness of a human variable cross-section bone based on ultrasonic guided waves, characterized in that, The method comprises: S100, acquiring an ultrasonic guided wave signal acting on human cortical bone; S200, preprocessing the ultrasonic guided wave signal to separate out a direct A0 mode signal and a direct S0 mode signal; S300, converting time-space domain signals of the direct A0 mode signal and the S0 mode signal into wave number space domain signals to obtain a spatial wave number distribution of the A0 mode and a spatial wave number distribution of the S0 mode; S400, obtaining a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode; S500, obtaining an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function; S600, substituting wave number space ridge curves at different angles into corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

2. The method of claim 1, wherein, In S200, the preprocessing comprises: performing band-pass filtering processing on the acquired ultrasonic guided wave signal according to a center frequency of an excitation narrowband signal.

3. The method according to claim 1 or 2, characterized in that, The S200 comprises the following steps: S201, performing band-pass filtering processing on the acquired ultrasonic guided wave signal according to a center frequency of an excitation narrowband signal; S202, obtaining a propagation time of the direct A0 mode signal and a propagation time of the direct S0 mode signal according to a propagation speed of the ultrasonic guided wave signal in the human cortical bone, and further cutting the filtered ultrasonic guided wave signal to obtain the direct A0 mode signal and the S0 mode signal.

4. The method of claim 1, wherein, In S300, fast Fourier transform is performed on the time-space domain signals of the direct A0 mode signal and the S0 mode signal to obtain frequency domain signals arranged according to spatial positions; then, the frequency of the signal is fixed as the frequency of the excitation signal, a short spatial Fourier transform is used in the spatial domain, and a spatial window function is translated on the multi-point ultrasonic guided wave signal to obtain spatial wave number distributions at different positions of the A0 mode and spatial wave number distributions at different positions of the S0 mode.

5. The method of claim 1, wherein, In S500, the wave number space ridge curve is substituted into the dispersion curve fitting function to obtain the inverted cortical bone thickness distribution rule.

6. An ultrasonic guided wave based human variable cross-section bone thickness detection system, characterized in that, The system comprises: a measurement module configured to acquire an ultrasonic guided wave signal acting on human cortical bone; a signal preprocessing module configured to preprocess the ultrasonic guided wave signal to separate out a direct A0 mode signal and a direct S0 mode signal; a wave number domain conversion module configured to convert time-space domain signals of the direct A0 mode signal and the S0 mode signal into wave number space domain signals to obtain a spatial wave number distribution of the A0 mode and a spatial wave number distribution of the S0 mode; a time-frequency ridge extraction module configured to obtain a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the A0 mode and the spatial wave number distribution of the S0 mode; a thickness inversion module configured to obtain an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function; a two-dimensional thickness imaging module configured to substitute wave number space ridge curves at different angles into corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

7. The system of claim 6, wherein, The signal preprocessing module comprises: The filtering module performs band-pass filtering on the obtained ultrasonic guided wave signal according to the center frequency of the excitation narrowband signal; The separation module is configured to intercept the direct Ao mode signal and the direct S0 mode signal from the filtered ultrasonic guided wave signal according to the propagation time.

8. The system of claim 6, wherein, The wave number domain conversion module comprises: The frequency domain conversion module is configured to obtain the preprocessed ultrasonic guided wave signal, perform fast Fourier transform on the multi-array ultrasonic guided wave signal, and obtain a frequency domain signal arranged according to spatial positions; The short spatial conversion module is configured to fix the frequency of the signal as the excitation frequency, use a short spatial Fourier transform in the spatial domain, and use a spatial window function to shift on the multi-point ultrasonic guided wave signal to obtain wave number power spectrums at different positions.

9. The system of claim 8, wherein, The system further comprises a processor configured to perform the steps of human variable cross-section bone thickness detection based on ultrasonic guided waves, including: preprocessing the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct S0 mode signal; converting the time-space domain signals of the direct Ao mode signal and the S0 mode signal into wave number space domain signals to obtain spatial wave number distribution of the Ao mode and spatial wave number distribution of the S0 mode; obtaining a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode; obtaining an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function; substituting the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

10. An ultrasonic guided wave based human variable cross-section bone thickness detection system, comprising: The system comprises a signal generator (101), a power amplifier (102), an excitation transducer (103), a receiving array transducer (104), and a computer device (105). The signal generator (101) is configured to transmit an ultrasonic guided wave signal to a human measured long bone, and excite a sinusoidal Hanning window amplitude modulation pulse signal with a narrowband frequency domain. The power amplifier (102) amplifies the energy of the pulse signal, and excites ultrasonic guided waves in the cortical bone of the long bone through the excitation transducer (103). The array transducer (104) receives the ultrasonic guided wave signal along the direction of the human measured long bone shaft, and transmits the signal to the computer device (105). The computer device (105) comprises a memory and a processor. The memory is configured to obtain a cortical bone two-dimensional thickness distribution image. The processor is configured to perform the following steps: preprocessing the ultrasonic guided wave signal to separate the direct Ao mode signal and the direct S0 mode signal; converting the time-space domain signals of the direct Ao mode signal and the S0 mode signal into wave number space domain signals to obtain spatial wave number distribution of the Ao mode and spatial wave number distribution of the S0 mode; obtaining a wave number space ridge curve by using a time-frequency ridge extraction algorithm according to the spatial wave number distribution of the Ao mode and the spatial wave number distribution of the S0 mode; obtaining an inverted cortical bone thickness distribution rule according to the wave number space ridge curve and a dispersion curve fitting function; substituting the wave number space ridge curves at different angles into the corresponding dispersion curve fitting functions to obtain a cortical bone two-dimensional thickness distribution image.

Images