Human osteoporosis detection method based on ultrasonic guided wave dispersion curve

Abstract

The present invention provides a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve, which is used for processing and analyzing the collected ultrasonic signal of human bones to obtain the corresponding bone density judgment basis, and is characterized in that it includes the following steps: Fourier transforming the ultrasonic signal to obtain the initial dispersion curve; using a predetermined power spectrum estimation algorithm to process the initial dispersion curve to obtain the maximum energy point trajectory; using a predetermined search algorithm to match the fitting points in a pre-established simulation model database for the maximum energy point trajectory, and obtaining the parameter of the maximum fitting coefficient corresponding to the maximum energy point trajectory, and using the shear wave velocity in the parameter as the bone density judgment basis so that the user can distinguish between non-osteoporotic and osteoporotic patients according to the bone density judgment basis. The present invention has no radiation hazard to the patient, and has low cost, high accuracy, small calculation amount, and high real-time performance.

Description

Technical Field

[0001] The invention belongs to the field of medical diagnosis, and in particular relates to a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves. Background Art

[0002] Osteoporosis is a common degenerative disease. Once osteoporosis occurs, it will bring many harms to patients, such as back pain, height loss, hunchback, dyspnea, osteoporotic fractures, etc. Therefore, early screening and early prevention of osteoporosis are the most economical and effective means to prevent and treat the occurrence of osteoporosis and osteoporotic fractures.

[0003] At present, most medical institutions at home and abroad use dual energy X-ray absorptiometry (DEXA) as a method to evaluate bone density to evaluate the health of bones. However, due to the complex operation of dual energy X-ray absorptiometry, long measurement time, high test cost, expensive test equipment and large space occupied, it is difficult to popularize and promote it in community medical institutions, small and medium-sized hospitals or physical examination centers. This also leads to a large number of people with osteoporosis not being effectively diagnosed and prevented, resulting in a high incidence of osteoporosis in middle-aged and elderly people. In addition, special populations such as premature infants or pregnant women often need to monitor their bone health regularly, and the low radiation hazard of DEXA is not suitable for long-term monitoring of pregnant women and infants. Summary of the invention

[0004] In order to solve the above problems, a human osteoporosis detection algorithm is provided which is free of radiation hazards, simple and easy to use, low in cost and high in accuracy. The present invention adopts the following technical solutions:

[0005] The present invention provides a human osteoporosis detection method based on ultrasonic guided wave dispersion curve, which is used for processing and analyzing the collected ultrasonic signal of human bones to obtain the corresponding bone density judgment basis, and is characterized in that it includes: step S1-1, Fourier transforming the ultrasonic signal to obtain an initial dispersion curve; step S1-2, processing the initial dispersion curve using a predetermined power spectrum estimation algorithm to obtain a maximum energy point trajectory; step S1-3, matching fitting points in a pre-established simulation model database using a predetermined search algorithm for the maximum energy point trajectory, obtaining a parameter of a maximum fitting coefficient corresponding to the maximum energy point trajectory, and using the shear wave velocity in the parameter as a bone density judgment basis so that a user can distinguish between non-osteoporotic and osteoporotic patients according to the bone density judgment basis.

[0006] A method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves provided by the present invention may also have such a technical feature, wherein the simulation model database is a dispersion curve point of a three-layer simulation of skin-cortical bone-bone marrow constructed by the guided wave signal collected by the waveguide.

[0007] According to a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve provided by the present invention, the method may also have such a technical feature, wherein the process of establishing a simulation model database comprises the following steps: Step S2-1, establishing an initial database with an initial cortical bone thickness of h1mm, a thickness increment of Δh, an initial longitudinal wave velocity of p1m / s, a longitudinal wave velocity increment of Δp, an initial shear wave velocity of s1m / s and a shear wave velocity increment of Δs, wherein the initial database includes h z *p z *s z Initial dispersion data K hps ; Step S2-2, reorganize all initial dispersion data into the corresponding dispersion point matrix K′ hps , thus obtaining the matrix K′ composed of all frequency dispersion points hps The simulation model database composed of the initial dispersion data K hps It is composed of multiple dispersion points. The initial dispersion data K hps Includes multiple modes Wherein, n is the number of modes corresponding to each set of initial dispersion data, m=(1, 2) is the frequency and phase velocity of the dispersion points in each mode, and l is the arrangement order of the dispersion points in each mode.

[0008] According to the human osteoporosis detection method based on ultrasonic guided wave dispersion curve provided by the present invention, it can also have such a technical feature, wherein step S2-2 includes the following sub-steps: step S2-2-1, based on the frequency point, searching for the corresponding phase velocity position b in each mode i :

[0009] b i =i

[0010]

[0011] Wherein, df is the longitudinal interval of the maximum energy point trajectory A, l is the arrangement order of the frequency dispersion points in each mode, and c is an array consisting of all phase velocity positions corresponding to the frequency points under each arrangement order l; step S2-2-2, set the arrangement order l corresponding to the minimum value in the array c as the known arrangement sequence l′; step S2-2-3, calculate the wave number position b according to the known arrangement sequence l′ j :

[0012]

[0013] Where dk is the lateral spacing of the maximum energy point trajectory A, is the frequency of the l'th dispersion point of the nth mode, is the phase velocity of the l'th dispersion point of the nth mode; step S2-2-4, according to the phase velocity position b i and wave number position b j The reorganized initial dispersion data is obtained as the dispersion point matrix K′ hps , and the initial dispersion data K hps Mark the corresponding position in the matrix to obtain the marked position; step S2-2-5, repeat steps S2-2-1 to S2-2-4 until all the initial dispersion data are reorganized to obtain the matrix K′ composed of all the dispersion points hps The simulation model database is constructed.

[0014] According to a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve provided by the present invention, the method may also have such a technical feature, wherein step S1-3 includes the following sub-steps: step S1-3-1, judging whether the energy E of the grid point in the maximum energy point trajectory is greater than a predetermined energy threshold T; step S1-3-2, when step S1-3-1 is judged to be yes, recording the position corresponding to the grid point as the position to be matched; step S1-3-3, searching in the simulation model database through a search algorithm whether there is a marked position corresponding to the position to be matched; step S1-3-4, when step S1-3-3 is judged to be yes, according to the energy E of the grid point corresponding to the marked position and the fitting point P hps Calculate the fitting coefficient Q hps And the mean value of the fitting points P' hps :

[0015] P hps =P hps +1

[0016]

[0017]

[0018] ; Step S1-3-5, repeat steps S1-3-1 to S1-3-4 until the fitting coefficient Q corresponding to each grid point in the maximum energy point trajectory is obtained hps And the mean value of the fitting points P' hps ; Step S1-3-6, according to all fitting coefficients Q hps And the mean value of all fitting points P' hps Select the fitting coefficient Q hps The maximum value is taken as the maximum fitting coefficient, and the shear wave velocity in the parameters of the maximum fitting coefficient is taken as the basis for determining the bone density.

[0019] A method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve provided by the present invention may also have such a technical feature, wherein the power spectrum estimation algorithm in step S1-2 is a Burg algorithm.

[0020] A method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves provided by the present invention may also have such a technical feature, wherein the search algorithm in step S1-3 is a grid search algorithm.

[0021] Function and Effect of the Invention

[0022] According to a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve of the present invention, the ultrasonic signal is first subjected to Fourier transformation to obtain an initial dispersion curve, and then the initial dispersion curve is processed using a predetermined power spectrum estimation algorithm to obtain a maximum energy point trajectory, and finally a predetermined search algorithm is used to match fitting points in a pre-established simulation model database for the maximum energy point trajectory, so as to obtain a parameter of a maximum fitting coefficient corresponding to the maximum energy point trajectory, and the shear wave velocity in the parameter is used as a basis for bone density judgment, so that in the case of only a patient's human bone ultrasonic signal, the cortical bone shear wave velocity can be inverted through the above steps, and the shear wave velocity can reflect the porosity of the cortical bone, and non-osteoporotic and osteoporotic patients can be effectively distinguished according to the shear wave velocity, with high accuracy and strong reliability. In addition, since only the patient's human bone ultrasonic signal is required, compared with the dual-energy X-ray absorption method, the method of the present invention is low in cost and does not generate radiation to the patient, and the present invention also has a small amount of calculation and high real-time performance, and is suitable for long-term and regular detection of the health of human bones. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves according to an embodiment of the present invention;

[0024] Figure 2 A grid diagram of the maximum energy point trajectory according to an embodiment of the present invention;

[0025] Figure 3 A grid diagram of initial dispersion data according to an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of a grid of a frequency dispersion point matrix according to an embodiment of the present invention;

[0027] Figure 5 Flow chart of sub-steps of step S1-3 of an embodiment of the present invention;

[0028] Figure 6A schematic diagram of grid points corresponding to effective energy values ​​according to an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of matching a marked position with a position to be matched according to an embodiment of the present invention;

[0030] Figure 8 A dispersion curve diagram obtained by inversion of fitting point coefficients according to an embodiment of the present invention;

[0031] Fig. 9 A dispersion curve diagram obtained by inversion based on the mean value of the fitting points according to an embodiment of the present invention; and

[0032] Fig.10 The dispersion curve diagram is obtained by inverting the mean value of the fitting points and the fitting coefficients according to the embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following is a detailed description of a human osteoporosis detection method based on ultrasonic guided wave dispersion curves of the present invention in combination with embodiments and drawings.

[0034] <Example>

[0035] In an embodiment of the present invention, a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves is to process and analyze the ultrasonic signal collected on the same side of the human radial bone to obtain the corresponding bone density judgment basis. The ultrasonic signal is obtained by a single-transmit multiple-receive method.

[0036] Figure 1 The present invention is a flowchart of a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curves according to an embodiment of the present invention.

[0037] like Figure 1 As shown, a method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve includes the following steps:

[0038] Step S1-1, performing Fourier transform on the ultrasonic signal to obtain an initial dispersion curve.

[0039] The ultrasonic signal related to human bones is multi-channel time-displacement format data, the Fourier transform is specifically a two-dimensional Fourier transform, and the initial dispersion curve is data in the frequency-wave number domain format.

[0040] In this embodiment, the ultrasonic signal related to human bones is a 60-channel time-displacement signal.

[0041] Step S1-2: using a predetermined power spectrum estimation algorithm to process the initial dispersion curve to obtain a maximum energy point trajectory.

[0042] The power spectrum estimation algorithm is the Burg algorithm.

[0043] Figure 2 Schematic diagram of a grid of the maximum energy point trajectory according to an embodiment of the present invention.

[0044] Specifically, the initial dispersion curve is processed using the Burg algorithm to obtain a high-resolution and high-precision maximum energy point trajectory A (such as Figure 2 As shown), the maximum energy point trajectory A is an information matrix in the frequency-wave number domain format. Among them, the horizontal axis is the frequency of the collected ultrasonic signal, and the number of columns is recorded as a i ; The ordinate is the wave number of the collected ultrasonic signal, and the row number is recorded as a j .

[0045] The maximum energy point trajectory A is the reference matrix, which is composed of various grid points. The position of each grid point is expressed as [a j , a i ], each grid point has a corresponding energy value E (such as Figure 2 E 11 、E 22 、E 33 、E 44 ...E ji ...).

[0046] Step S1-3, using a predetermined search algorithm to match fitting points in a pre-established simulation model database for the maximum energy point trajectory, obtain the parameters of the maximum fitting coefficient corresponding to the maximum energy point trajectory, and use the shear wave velocity in the parameters as a basis for bone density judgment, so that users can distinguish between non-osteoporotic and osteoporotic patients based on the bone density judgment basis.

[0047] The search algorithm in step S1-3 is a grid search algorithm.

[0048] The simulation model database is a dispersion curve point of the skin-cortex-bone-bone marrow three-layer simulation constructed by the waveguide signal collected by the waveguide. The process of establishing the database includes the following steps:

[0049] Step S2-1, establish an initial database with an initial cortical bone thickness of h1mm, a thickness increment of Δh, an initial longitudinal wave velocity of p1m / s, a longitudinal wave velocity increment of Δp, an initial shear wave velocity of s1m / s and a shear wave velocity increment of Δs, wherein the initial database includes h z *p z *s z Initial dispersion data K hps .

[0050] Among them, h1, Δh, p1, Δp, s1 and Δs are all positive numbers. Frequency dispersion matrix Khps It is expressed as dispersion data when the thickness is h, the longitudinal wave velocity is p, and the shear wave velocity is s.

[0051] Each group of initial dispersion data K hps It is composed of multiple dispersion points, and each has multiple modes (A0, A1, A2...S0, S1, S2...). All modes of this group of initial dispersion data are marked as Wherein, n is the number of modes in each set of data, m=(1, 2) are the frequency and phase velocity of the dispersion points of each mode, and l is the arrangement order of the dispersion points of each mode.

[0052] Figure 3 Schematic diagram of a grid of initial dispersion data according to an embodiment of the present invention.

[0053] like Figure 3 As shown, the initial dispersion data is represented by a matrix, the horizontal axis of which is the frequency of the ultrasonic signal in the database, and the number of columns is recorded as b i ; The ordinate is the wave number of the ultrasonic signal in the database, and the row number is recorded as b j .

[0054] Step S2-2, reorganize all initial dispersion data into the corresponding dispersion point matrix K′ hps , thus obtaining the matrix K′ composed of all frequency dispersion points hps The simulation model database is constructed.

[0055] Wherein, step S2-2 includes the following sub-steps:

[0056] Step S2-2-1, search for the corresponding phase velocity position b in each mode based on the frequency point i :

[0057]

[0058] Where df is the longitudinal interval of the maximum energy point trajectory A, l is the arrangement order of the dispersion points in each mode, and c is an array of all phase velocity positions corresponding to the frequency points in each arrangement order l.

[0059] Step S2-2-2, set the arrangement sequence l corresponding to the minimum value in array c as the known arrangement sequence l′.

[0060] Step S2-2-3, calculate the wave number position b of the wave number according to the known arrangement sequence l′ j :

[0061]

[0062] Where dk is the lateral spacing of the maximum energy point trajectory A, is the frequency of the l'th dispersion point of the nth mode, is the phase velocity of the l'th dispersion point of the nth mode.

[0063] Figure 4 Schematic diagram of a grid of a frequency dispersion point matrix according to an embodiment of the present invention.

[0064] Step S2-2-4, according to the phase velocity position b i and wave number position b j The reorganized initial dispersion data is obtained as the dispersion point matrix K′ hps (like Figure 4 As shown), and the initial dispersion data K hps Mark the corresponding position in the mark position R i :

[0065] R i =K(b j , b i )=1 (3)

[0066] Step S2-2-5, repeating steps S2-2-1 to S2-2-4 until all initial dispersion data are reorganized to obtain a matrix K′ consisting of all dispersion points hps The simulation model database is constructed.

[0067] Figure 5 Flow chart of sub-steps of step S1-3 of an embodiment of the present invention.

[0068] like Figure 5 As shown, step S1-3 includes the following sub-steps:

[0069] Step S1-3-1, determine whether the energy E of the grid point in the maximum energy point trajectory is greater than the predetermined energy threshold T. If it is determined to be yes, proceed to step S1-3-2. If it is determined to be no, repeat step S1-3-1 to determine the next grid point.

[0070] Figure 6 It is a schematic diagram of grid points corresponding to effective energy values ​​according to an embodiment of the present invention.

[0071] Due to the complexity of the ultrasonic signals collected from human bones, when extracting the maximum energy point trajectory of the ultrasonic dispersion curve of the lamb wave in the human radial bone, an energy threshold T is pre-set for the energy matrix value of the dispersion curve. When the energy E is greater than the energy threshold T, the energy E is taken as the effective energy value (such as Figure 6 shown).

[0072] Step S1-3-2: if the judgment in step S1-3-1 is yes, record the position corresponding to the grid point as the position to be matched Δa i .

[0073] Step S1-3-3, through the search algorithm, search in the simulation model database whether there is a marked position corresponding to the position to be matched, if it is judged to be yes, go to step S1-3-4, if it is judged to be no, go to step S1-3-1.

[0074] Figure 7 It is a schematic diagram of matching the marked position and the position to be matched according to an embodiment of the present invention.

[0075] like Figure 7 As shown, in the frequency dispersion matrix K′ hps The position to be matched in i Search all b i , determine at the position to be matched Δa i Whether a marked grid point (i.e., the marked position R in step S2-2-4) is found i ).

[0076] Step S1-3-4: when the judgment in step S1-3-3 is yes, the energy E of the grid point corresponding to the marked position and the fitting point P hps Calculate the fitting coefficient Q hps And the mean value of the fitting points P' hps :

[0077]

[0078] Specifically, when the judgment in step S1-3-3 is yes, each mark position R i The energy value E of the grid point at i Accumulate all b i The corresponding energy is taken as the effective energy E′:

[0079] E′=E+E i (5)

[0080] At the same time, the mark position R i The fitting point P at hps For all b i The energy value and fitting points are accumulated and the fitting coefficient Q is calculated according to Formula 5 hps , and calculate the mean value of the fitting point P' according to formula 5 hps :

[0081] Step S1-3-5, repeat steps S1-3-1 to S1-3-4 until the fitting coefficient Q corresponding to each grid point in the maximum energy point trajectory is obtained hps And the mean value of the fitting points P' hps .

[0082] Step S1-3-6, according to all fitting coefficients Q hps And the mean value of all fitting points P' hps Select the fitting coefficient Q hps The maximum value is taken as the maximum fitting coefficient, and the shear wave velocity in the parameters of the maximum fitting coefficient is taken as the basis for determining the bone density.

[0083] Figure 8 This is a dispersion curve diagram obtained based on the inversion of the fitting point coefficients according to an embodiment of the present invention.

[0084] like Figure 8 As shown, when selecting the fitting coefficient Q hps In the process of maximum value, if only the fitting point coefficient Q is considered hps Without considering the mean value of the fitting points P' hps , then the maximum fitting point coefficient Q hps Relatively large, but the mean of the fitting points P' hps Smaller.

[0085] Fig. 9 This is a dispersion curve diagram obtained based on fitting point mean inversion according to an embodiment of the present invention.

[0086] If only the mean of the fitting points P' is considered hps Without considering the fitting point coefficient Q hps , then the mean of the fitting points is P' hps Relatively large, but the maximum fitting point coefficient Q hps Smaller (such as Fig. 9 shown).

[0087] Therefore, based only on the fitting coefficient Q hps Or just based on the mean of the fitting points P' hps Selecting the maximum fitting coefficient will lead to inaccurate matching results, so the fitting coefficient Q needs to be considered comprehensively. hps And the mean value of the fitting points P' hps .

[0088] Fig.10 The dispersion curve diagram is obtained by inverting the mean value of the fitting points and the fitting coefficients according to the embodiment of the present invention.

[0089] like Fig.10 As shown, the mean value P' for all fitting points in the simulation model database is hps For comparison, only The fitting coefficient Q when listening to the corresponding thickness h, longitudinal wave velocity p, and shear wave velocity s hps Then, in these fitting coefficients Q hps Select Q hps，max (i.e. the maximum fitting coefficient), at this time Q hps，maxThe corresponding thickness h, longitudinal wave velocity p, and shear wave velocity s are the best inversion parameters. When the shear wave velocity is lower than the reference value, the patient can be judged to have osteoporosis.

[0090] Example Function and Effect

[0091] According to the human osteoporosis detection method based on ultrasonic guided wave dispersion curve provided by this embodiment, the ultrasonic signal is first Fourier transformed to obtain the initial dispersion curve, and then the initial dispersion curve is processed by a predetermined power spectrum estimation algorithm to obtain the maximum energy point trajectory, and finally the predetermined search algorithm is used to match the fitting points in the pre-established simulation model database for the maximum energy point trajectory, and the parameter of the maximum fitting coefficient corresponding to the maximum energy point trajectory is obtained, and the shear wave velocity in the parameter is used as the basis for bone density judgment. Therefore, in the case of only the patient's human bone ultrasonic signal, the cortical bone shear wave velocity can be inverted through the above steps. The shear wave velocity can reflect the porosity of the cortical bone. According to the shear wave velocity, non-osteoporotic and osteoporotic patients can be effectively distinguished, and it has high accuracy and strong reliability. In addition, since only the patient's human bone ultrasonic signal is required, compared with the dual-energy X-ray absorption method, the method of the present invention is low in cost and does not generate radiation to the patient. At the same time, the present invention also has a small amount of calculation and high real-time performance, and is suitable for long-term and regular detection of the health of human bones.

[0092] In addition, in the embodiment, since all the initial dispersion data are reorganized into the corresponding dispersion point matrix K′ hps , so in the frequency dispersion matrix K′ hps In the case of the same format as the maximum energy point trajectory, the present invention obtains the parameters of the maximum fitting coefficient corresponding to the maximum energy point trajectory more quickly in the simulation model database through a search algorithm.

[0093] The above embodiments are only used to illustrate specific implementation modes of the present invention, and the present invention is not limited to the description scope of the above embodiments.

Claims

1. A method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve, which is used to process and analyze the collected ultrasonic signals of human bones to obtain the corresponding bone density judgment basis, characterized in that: The steps include: Step S1-1, performing Fourier transform on the ultrasonic signal to obtain an initial dispersion curve; Step S1-2, using a predetermined power spectrum estimation algorithm to process the initial dispersion curve to obtain a maximum energy point trajectory; Step S1-3, using a predetermined search algorithm to match fitting points in a pre-established simulation model database for the maximum energy point trajectory, obtaining a parameter of a maximum fitting coefficient corresponding to the maximum energy point trajectory, and using the shear wave velocity in the parameter as a basis for determining bone density so that the user can distinguish between non-osteoporotic and osteoporotic patients based on the bone density determination basis. The simulation model database is a dispersion curve point of the skin-cortex-bone-bone marrow three-layer simulation constructed by the waveguide signal collected by the waveguide. The establishment process of the simulation model database includes the following steps: Step S2-1, establish an initial database with an initial cortical bone thickness of h1mm, a thickness increment of Δh, an initial longitudinal wave velocity of p1m / s, a longitudinal wave velocity increment of Δp, an initial shear wave velocity of s1m / s and a shear wave velocity increment of Δs, wherein the initial database includes h z *p z *s z Initial dispersion data K hps ; Step S2-2: reorganize all the initial dispersion data into a corresponding dispersion point matrix K′ hps , thus obtaining the matrix K′ composed of all the frequency dispersion points hps The simulation model database constituted by The initial dispersion data K hps It is composed of multiple dispersion points. The initial dispersion data K hps Includes multiple modes Wherein, n is the number of the modes corresponding to each set of the initial dispersion data, m=(1,2) is the frequency and phase velocity of the dispersion points under each mode, l is the arrangement order of the dispersion points under each mode, The step S2-2 includes the following sub-steps: Step S2-2-1, searching for the corresponding phase velocity position b in each of the modes based on the frequency point i : b i =i Wherein, df is the longitudinal interval of the maximum energy point trajectory A, l is the arrangement order of the frequency dispersion points in each mode, and c is an array consisting of all the phase velocity positions corresponding to the frequency points in each arrangement order l; Step S2-2-2, setting the arrangement sequence l corresponding to the minimum value in the array c as the known arrangement sequence l′; Step S2-2-3, calculating the wave number position b of the wave number according to the known arrangement sequence l′ j : Where dk is the lateral spacing of the maximum energy point trajectory A, is the frequency of the l'th dispersion point of the nth mode, is the phase velocity of the l'th dispersion point of the nth mode; Step S2-2-4, according to the phase velocity position b i And the wave number position b j The reorganized initial dispersion data is obtained as the dispersion point matrix K′ hps , and the initial dispersion data K hps Mark the corresponding position in to obtain the marked position; Step S2-2-5, repeating the steps S2-2-1 to S2-2-4 until all the initial dispersion data are reorganized to obtain a matrix K′ composed of all the dispersion points hps The simulation model database constituted by The step S1-3 includes the following sub-steps: Step S1-3-1, determining whether the energy E of the grid point in the maximum energy point trajectory is greater than a predetermined energy threshold T; Step S1-3-2, when the judgment in step S1-3-1 is yes, recording the position corresponding to the grid point as the position to be matched; Step S1-3-3, searching the simulation model database by the search algorithm to see whether there is the marked position corresponding to the position to be matched; Step S1-3-4, when the judgment in step S1-3-3 is yes, according to the energy E of the grid point corresponding to the marked position and the fitting point P hps Calculate the fitting coefficient Q hps And the mean value of the fitting points P' hps : Step S1-3-5, repeating steps S1-3-1 to S1-3-4 until the fitting coefficient Q corresponding to each grid point in the maximum energy point trajectory is obtained hps And the mean value of the fitting points P' hps ; Step S1-3-6, according to all the fitting coefficients Q hps And the mean value P' of all the fitting points hps Select the fitting coefficient Q hps The maximum value is taken as the maximum fitting coefficient, and the shear wave velocity in the parameters of the maximum fitting coefficient is taken as the basis for determining the bone density.

2. The method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve according to claim 1, characterized in that: in, The power spectrum estimation algorithm in step S1-2 is the Burg algorithm.

3. The method for detecting human osteoporosis based on ultrasonic guided wave dispersion curve according to claim 1, characterized in that: in, The search algorithm in step S1-3 is a grid search algorithm.

Images