KR102364309B1 - Method for estimating bone mineral density and bone structure using ultrasonic attenuation coefficient and backscatter coefficient - Google Patents

Abstract

A first measurement step of irradiating a transmitting ultrasound wave to a measurement target by any one transducer, and receiving a first receiving ultrasound wave transmitted through the measurement target by another transducer facing the one transducer; a second measurement step of irradiating a transmission ultrasound wave toward the measurement object by any one transducer, and receiving a second reception ultrasound wave returned by reflecting the measurement object; an extraction step of extracting an attenuation coefficient and a backscattering coefficient for a frequency function by analyzing the electrical signals of the first ultrasonic wave and the second ultrasonic wave received; and applying the attenuation coefficient and backscattering coefficient extracted in the extraction step as independent variables to a multiple regression model in which any one of bone density and bone structure of the measurement object is a dependent variable, and at least two or more variables are independent variables. A prediction step of predicting the bone density and bone structure of the subject; It provides a method for predicting bone density and bone structure using an ultrasonic damping coefficient and a backscattering coefficient, characterized in that it comprises a.

Description

Bone density and bone structure prediction method using ultrasound attenuation coefficient and backscatter coefficient {METHOD FOR ESTIMATING BONE MINERAL DENSITY AND BONE STRUCTURE USING ULTRASONIC ATTENUATION COEFFICIENT AND BACKSCATTER COEFFICIENT}

The present invention relates to a method for predicting bone density and bone structure using an ultrasonic damping coefficient and a backscattering coefficient.

Osteoporosis is defined as a systemic bone disease in which fractures easily occur even with a small impact due to a decrease in bone mass and destruction of bone structure. At this time, early diagnosis and prevention are important in that osteoporosis is an irreversible disease that cannot be returned to a normal state once it is onset.

Among various diagnostic methods for diagnosing osteoporosis, there is a representative method for measuring bone density using dual energy X-ray absorptiometry (DEXA). Because the measurement must be made, the dual energy X-ray absorptiometry using radiation has a risk of repeated exposure to radiation, and in particular, there is a difficulty in measuring the bone density of pregnant women.

Accordingly, research on a technique for diagnosing osteoporosis using a method harmless to the human body is being conducted from various angles, and as a part of this technique, a quantitative ultrasound (QUS) technique for diagnosing osteoporosis using ultrasound has been proposed.

Here, the quantitative ultrasound technology irradiates ultrasound to the heel bone of the human body, that is, the calcaneus, and then measures ultrasound characteristics such as the speed of sound (SOS) and broadband ultrasound attenuation (BUA) of the calcaneus. Thus, the bone mineral density (BMD) of the whole body can be predicted indirectly, and osteoporosis can be diagnosed from this.

This quantitative ultrasound technique has the advantage that there is no risk of radiation exposure even for repeated examinations, and the price and examination cost of the diagnostic device are relatively low compared to the osteoporosis diagnosis method using radiation. The technology related to this is disclosed in Korean Patent Application Laid-Open No. 10-2005-0038812 (application date: October 23, 2003, publication date: April 29, 2005), and Korean Patent Application Laid-Open No. 10-2003-0034550 (application date: 2001. 10. 26., Published date: 2004. 06. 11.), etc. have been suggested.

On the other hand, even with the same bone density, there is a need to consider changes in bone structure as well as bone density because the incidence of fractures varies depending on age, fracture-related history, steroid drug treatment, and differences in bone structure of the human body.

However, since the conventional quantitative ultrasound technology measures only the change in bone density for the diagnosis of osteoporosis, it is difficult to identify and diagnose the cause of osteoporosis caused by the change in bone structure.

An object of the present invention is to provide a technique capable of improving the accuracy of diagnosis of osteoporosis by evaluating bone density and bone structure together through ultrasound measurement to solve the above problems.

In order to achieve this object, in a method for predicting bone density and bone structure using an ultrasonic damping coefficient and a backscattering coefficient according to an embodiment of the present invention, any one transducer irradiates a transmission ultrasound to a measurement object, and the any one of the transducers a first measuring step of receiving a first receiving ultrasonic wave that has passed through the measurement target by another transducer facing the transducer; a second measurement step of irradiating a transmission ultrasound wave toward the measurement object by any one transducer, and receiving a second reception ultrasound wave returned by reflecting the measurement object; an extraction step of extracting an attenuation coefficient and a backscattering coefficient for a frequency function by analyzing the electrical signals of the first ultrasonic wave and the second ultrasonic wave received; and applying the attenuation coefficient and backscattering coefficient extracted in the extraction step as independent variables to a multiple regression model in which any one of bone density and bone structure of the measurement object is a dependent variable, and at least two or more variables are independent variables. A prediction step of predicting the bone density and bone structure of the subject; may include

Here, in the extraction step, an attenuation coefficient for a frequency function in a frequency band of 0.2 MHz to 0.6 MHz (hereinafter referred to as a 'calculation frequency band') through the electrical signals of the first and second reception ultrasound waves and the rear calculating a scattering coefficient; and a variable extraction step of extracting an attenuation coefficient and a backscattering coefficient corresponding to a center band from the calculated frequency band.

In this case, in the calculation step, the attenuation coefficient may be calculated by analyzing the first received ultrasonic wave, and may be determined by Equation 1 below.

[Equation 1]

[dB/cm]는 초음파 감쇠계수, d는 측정 대상체의 두께,

및

는 마주하는 한 쌍의 트랜스듀서 사이에 측정 대상체가 없는 경우와 있는 경우에 각각 수신된 신호의 파워스펙트럼레벨)(here,

[dB/cm] is the ultrasonic attenuation coefficient, d is the thickness of the measurement object,

and

is the power spectrum level of the received signal when there is no measurement object between a pair of transducers facing each other)

In addition, in the calculation step, the backscattering coefficient may be calculated by analyzing the first received ultrasound and the second received ultrasound, and may be determined by Equation 2 below.

[Equation 2]

[cm^-1sr^-1]는 초음파 후방산란계수, R은 어느 하나의 트랜스듀서의 초음파 송수신면으로부터 측정 대상체의 중앙부까지의 거리,

와

는 각각 측정 대상체 및 측정 대상체의 중앙부에 대응하는 거리에 위치된 완전 반사체(스테인리스 스틸)로부터 후방 산란된 신호의 파워스펙트럼레벨, W(f)는 초음파 초점에서 -3 dB만큼의 빔폭(반치 빔폭; half-maximum beam width),

는 제2 수신초음파 중 측정 대상체 내부에서 후방 산란된 초음파 신호가 차지하는 시간, l은 측정 대상체의 중앙부로부터 표면까지의 거리,

및

는 각각 투과법에 의해 측정된 측정 대상체의 위상속도 및 감쇠계수) (here,

[cm ^-1 sr ^-1 ] is the ultrasonic backscattering coefficient, R is the distance from the ultrasonic transmission/reception surface of any one transducer to the center of the measurement object,

Wow

is the power spectrum level of the signal backscattered from the perfect reflector (stainless steel) located at a distance corresponding to the measurement object and the center of the measurement object, respectively, W(f) is the beam width by -3 dB at the ultrasonic focus (half full beam width; half-maximum beam width),

is the time occupied by the ultrasonic signal backscattered inside the measurement object among the second reception ultrasound waves , l is the distance from the center of the measurement object to the surface,

and

is the phase velocity and attenuation coefficient of the measurement object measured by the transmission method, respectively)

In addition, the bone structure of the measurement target may include at least one of a bone volume ratio (BT/TV), a trabecular trabecular thickness (Tb.Th), and a trabecular trabecular interval (Tb.Sp).

In addition, the attenuation coefficient and the backscattering coefficient measured in advance from a plurality of measurement subject samples are used as independent variables, and any one of bone density and bone structure of the measurement object sample is used as a dependent variable, and a plurality of independent variables and each dependent variable are used. a multiple regression model generation step of performing multiple regression analysis of , and obtaining an index equation derived therefrom as the multiple regression model; may include

At this time, in the multiple regression model generation step, the multiple regression model is obtained from Equation 3 below, and the coefficient of determination (R ² , square of the Pearson correlation coefficient R) of the multiple regression model is determined between a single independent variable and a dependent variable. It can have a larger value than the coefficient of determination of the linear regression model.

[Equation 3]

및

는 독립변수(감쇠계수 및 후방산란계수), a 및 b는 계수, c는 y절편)(where y is the dependent variable (bone density or bone structure),

and

is the independent variable (attenuation coefficient and backscatter coefficient), a and b are coefficients, c is the y-intercept)

And, an output step of outputting the bone density and bone structure of the measurement object predicted in the prediction step; may further include.

As described above, according to the present invention, there are the following effects.

First, predicting bone density and bone structure together through a multiple regression model using two ultrasound variables (ultrasonic attenuation coefficient and backscattering coefficient) that are closely related to each other in addition to the parameters used in the existing quantitative ultrasound technology as independent variables Therefore, it is possible to improve the accuracy of diagnosis of osteoporosis compared to the conventional method of measuring bone density using only sound velocity and broadband ultrasound attenuation.

Second, in the present invention, by evaluating bone density and bone structure together, it is possible to identify and diagnose the cause of osteoporosis caused by changes in bone structure as well as changes in bone density.

Third, since the present invention diagnoses osteoporosis by irradiating ultrasound, it is possible to provide a safe measuring method to the measurer because there is no risk of the measurer being exposed to radiation.

1 shows an apparatus for predicting bone density and bone structure using an ultrasonic damping coefficient and a back scattering coefficient according to an embodiment of the present invention.
2 is a cross-sectional view illustrating a bone structure.
3 is a flowchart illustrating a method for predicting bone density and bone structure using an ultrasonic damping coefficient and a back scattering coefficient according to an embodiment of the present invention.
4 is a diagram of a frequency function calculated in an extraction step according to an embodiment of the present invention; It is a graph showing (a) attenuation coefficient and (b) backscattering coefficient.
5 is an illustration to explain the reliability of bone density and bone structure predicted from a multiple regression model to which an attenuation coefficient and a posterior coefficient are applied in predicting the bone density and bone structure of a measurement object, and a linear regression model and a multiple regression model This is a table comparing the coefficients of determination.

A preferred embodiment of the present invention will be described in more detail with reference to the accompanying drawings, but already known technical parts will be omitted or compressed for the sake of brevity of description.

<A brief description of the apparatus for predicting bone density and bone structure using ultrasonic damping coefficient and backscattering coefficient>

1 shows an apparatus for predicting bone density and bone structure (hereinafter, referred to as 'bone density and bone structure prediction apparatus') using an ultrasonic damping coefficient and a backscattering coefficient according to an embodiment of the present invention, and FIG. 2 is a bone structure is a cross-sectional view illustrating

Referring to FIG. 1 , the apparatus 100 for predicting bone density and bone structure according to an embodiment of the present invention includes a transducer 110 , a measurement unit 120 , an extraction unit 130 , a prediction unit 140 , and an output unit. (150) is included.

The transducer 110 may convert an applied electric signal into an ultrasonic wave or convert a received ultrasonic wave into an electric signal. At this time, the apparatus for predicting bone density and bone structure of the present invention is the through-transmission method shown in Figure 1 (a) and the pulse-echo method shown in Figure 1 (b) using the pulse-echo method. Ultrasound parameters for the measurement object M may be obtained.

More specifically, when the measurement object M is measured using the transmission method, any one of the pair of transducers 110 converts an electrical signal into an ultrasonic wave and transmits a transmission ultrasonic wave. , and the other transducer operates as a receiving unit that receives the first receiving ultrasonic wave passing through the measurement object M and converts it into an electrical signal. Here, the pair of transducers 110 are positioned so that one surface for transmitting and receiving ultrasound waves is opposite to each other, spaced apart by a predetermined distance between the pair of transducers 110 , and a measurement object between the pair of transducers 110 . (M) may be located.

When the measurement object M is measured using the pulse-echo method, one transducer 110 operates as a transmitter and a receiver, and the measurement object M is spaced apart from one surface of the transducer 110 by a certain distance. It is possible to irradiate a transmission ultrasound wave to the , and receive a second reception ultrasound wave returned by reflecting the measurement object (M).

For reference, the measurement using the pulse-echo method may be set and performed separately from the transmission method. In some cases, the measurement is performed after removing one transducer 110 in Fig. 1 (a) set to perform the transmission method, or any one transducer ( 110) may be driven to perform the measurement. In the latter case, the reflected signal returned by reflecting the other transducer 110 may also be included in the second receiving ultrasound, so that a separate signal processing for filtering the unnecessary signal for calculating the backscattering coefficient must be performed.

Here, the measurement object M may be a part of the body for which the bone density and bone structure are to be determined. More specifically, in addition to the calcaneus, it may be a region where osteoporotic fractures commonly occur, such as a lumbar vertebra, a femur, etc., but is not limited thereto. In this case, in an embodiment, the measurement target may be a bone, and more specifically, may be applied as a trabecular bone (T). For reference, in the apparatus 100 for predicting bone density and bone structure according to an embodiment of the present invention, a measurement object ( As M), a cancellous bone (T) sample prepared using the proximal portion of a bovine femur was used.

For reference, spongy bone (T) is a bone tissue constituting the bones of the human body together with cortical bone (C) having a very dense structure as shown in FIG. 2, and has a larger surface area than cortical bone (C). Since most of the bone tissue metabolism occurs in the cancellous bone (T), the bone density of the cancellous bone (T) is currently measured as an index for diagnosing osteoporosis in clinical practice.

And, the transducer 110 of the apparatus 100 for predicting bone density and bone structure according to an embodiment of the present invention may be installed in water. In this case, water has acoustic properties similar to those of soft tissues of the human body, and is used as an ultrasonic wave propagation medium. For example, the water may be distilled water.

The measuring unit 120 generates an electrical signal for the transmitted ultrasonic wave and provides a waveform generating part (not shown) to the transducer 110 and a collecting part for collecting the electrical signal of the receiving ultrasonic wave received by the transducer 110 ( (not shown) may be included. At this time, the transmission ultrasound generated by the waveform generating part may be in the form of a pulse or continuous wave, but in an embodiment of the present invention, it is preferable to generate the transmission ultrasound in the form of a pulse.

The extraction unit 130 extracts the attenuation coefficient and the backscattering coefficient for the frequency function by analyzing the electrical signals of the first and second reception ultrasound waves obtained by the measurement unit 120 . In this case, the extraction unit 130 may be configured to include a calculation part (not shown) and a variable extraction part (not shown).

Here, the calculation part may calculate the attenuation coefficient and backscattering coefficient of the 0.2 MHz to 0.6 MHz frequency band (hereinafter, referred to as 'calculated frequency band') by signal processing the electrical signals of the first and second reception ultrasound waves. . For reference, in the process of signal processing the electrical signal of the received ultrasonic wave, amplification and filtering of the electrical signal may be additionally performed.

And, the variable extraction part may extract the attenuation coefficient and the backscattering coefficient corresponding to the center band in the calculation frequency band. In this case, the center band may be 0.4 MHz, but is not limited thereto, and may vary according to a frequency band included in the calculation frequency band.

For reference, the transducer 110 in the apparatus for predicting bone density and bone structure according to an embodiment of the present invention is a non-focused transducer having a center frequency of 0.5 MHz, but is not limited thereto. In this case, the calculated frequency band may be varied according to the center frequency of the transducer 110 or the frequency band of the ultrasonic wave generated by the waveform generating part.

The prediction unit 140 is a multiple regression model in which any one of the bone density and bone structure of the measurement object M is a dependent variable, and at least two or more variables are independent variables. By applying the number as an independent variable, the bone density and bone structure of the measurement object (M) are predicted.

At this time, referring to FIG. 2 , the bone structure of the measurement target M is a bone volume ratio expressed as a ratio of the bone volume occupied by the bone trabecular Tb to the tissue volume (bone volume/tissue volume, BT/TV), and the trabecular thickness (trabecular) thickness; Tb.Th) and at least one of trabecular spacing (Tb.Sp), which means the interval between trabecular trabeculae (Tb).

In this case, the multiple regression model uses the attenuation coefficient and backscatter coefficient measured in advance from a plurality of measurement object (M) samples as independent variables, and uses any one of the bone density and bone structure of the measurement object (M) sample as a dependent variable. , multiple regression analysis of a plurality of independent variables and each dependent variable is performed, and an index equation derived therefrom may be pre-stored in the prediction unit 140 .

That is, the attenuation coefficient and backscattering coefficient, which are ultrasound variables extracted through ultrasound measurement, predict bone density and bone structure corresponding to the variable values through the multiple regression model pre-stored in the prediction unit 140, and from this, osteoporosis diagnosis result can be derived.

For reference, a detailed description of the prediction of bone density and bone structure of the measurement object M made through the above-described parts will be referred to in the method for predicting bone density and bone structure using ultrasound attenuation coefficients and backscattering coefficients to be described later.

The output unit 150 outputs the bone density and bone structure predicted from the prediction unit 140 . In this case, the output unit 150 may output the osteoporosis diagnosis result that can be derived from the bone density and bone structure of the measurement object M predicted through ultrasound measurement. In this case, the output unit 150 may be provided as a display monitor, a sound reproducing device, a printing device, etc. to provide the above-described output items as at least one of an image, a sound, and an output.

<Description of the method for predicting bone density and bone structure using ultrasonic damping coefficient and phase velocity>

3 is a flowchart illustrating a method for predicting bone density and bone structure using an ultrasonic damping coefficient and a backscattering coefficient according to an embodiment of the present invention, and FIG. 4 is a frequency function calculated in the extraction step according to an embodiment of the present invention. for (a) is a graph showing the attenuation coefficient and (b) the back scattering coefficient, Figure 5 is the bone density and bone structure predicted from a multiple regression model to which the attenuation coefficient and the posterior coefficient are applied in predicting the bone density and bone structure of the measurement object. As an illustration to explain the reliability of

1. First measurement step <S100>

In the first measurement step (S100), one transducer 110 irradiates a transmission ultrasonic wave to the measurement object located between the pair of transducers 110, and the other transducer 110 is the measurement object M ) is a step of receiving the first receiving ultrasound transmitted through. That is, the first measurement step ( S100 ) is a step of acquiring a first reception ultrasound wave that has passed through the measurement object M using a transmission method, as shown in FIG. 1A .

More specifically, the transmission ultrasound may be provided in the form of a pulse, and when the waveform generating part (not shown) of the measuring unit 110 provides an electrical signal for the pulse wave to any one of the transducers 110 , , the transducer 110 converts the received electrical signal into an ultrasonic signal to generate a pulse-shaped transmission ultrasound wave, and the generated pulse-shaped transmission ultrasound wave passes through the measurement object (M). Thereafter, the ultrasonic wave passing through the measurement object M is received by another transducer 110 , and the transducer 110 converts the received first ultrasonic wave into an electrical signal.

In this case, the pair of transducers 110 are positioned so that one surface for transmitting and receiving ultrasound waves faces, and a distance therebetween may be determined by a near field of the transducer 110 . Here, in the method for predicting bone density and bone structure using the ultrasonic attenuation coefficient and the backscattering coefficient according to an embodiment of the present invention, the transducer 110 is a non-focused transducer having a diameter of 25.4 mm and a center frequency of 0.5 MHz. However, the present invention is not limited thereto.

For reference, the near field means the point at which the beam width of the ultrasound is the narrowest when ultrasound is generated through a non-focused transducer, and the measurement object M is located at a point whose center coincides with the near field length (NFD). can be located

2. Second measurement step <S200>

The second measurement step (S200) is a step in which any one of the transducers 100 irradiates a transmission ultrasonic wave toward the measurement object M, and receives a second reception ultrasound wave returned by reflecting the measurement object M. That is, the second measurement step (S200) is a step of acquiring a second reception ultrasound wave returned by reflecting the measurement object M using the pulse-echo method, as shown in FIG. 1(b).

More specifically, the transmission ultrasound may be provided in the form of a pulse, and when the waveform generating part of the measuring unit 110 provides an electrical signal for the pulse wave to one transducer 110, the transducer ( 110) converts the received electrical signal into an ultrasonic signal to generate a pulse-shaped transmission ultrasound wave, and the generated pulse-shaped transmission ultrasound wave transmits or reflects toward the measurement object M. In this case, the ultrasonic wave reflected from the measurement object M is received by the transducer 110 , and the transducer 110 converts the received second ultrasonic wave into an electrical signal.

In this case, the measurement object M may be located at a point whose center is spaced apart by a near-field length (NFD) from one surface that transmits and receives ultrasound of the transducer 110 .

For reference, although the first measuring step (S100) and the second measuring step (S200) have been sequentially described for convenience of explanation, to summarize the above description, the first measuring step (S100) is the first measurement step (S100) using the transmission method. 1 is a step of obtaining a receiving ultrasound, and the second measuring step (S200) is a step of obtaining a second receiving ultrasound using a pulse-echo method, so in some cases, the second measuring step (S200) is a first measuring step ( S100) may be performed earlier.

3. Extraction step <S300>

The extraction step (S300) is a step of extracting an attenuation coefficient and a backscattering coefficient for a frequency function by analyzing the electrical signals of the first receiving ultrasound and the second receiving ultrasound.

Here, the extraction step (S300) is an attenuation coefficient and a backscatter meter for a frequency function in a frequency band of 0.2 MHz to 0.6 MHz (hereinafter referred to as a 'calculation frequency band') through the electrical signals of the first receiving ultrasonic wave and the second receiving ultrasonic wave. It may include a calculation step (S310) of calculating the number, and a variable extraction step (S320) of extracting an attenuation coefficient and a backscattering coefficient corresponding to the center band from the calculated frequency band.

In this case, the attenuation coefficient in the operation step S310 may be calculated by analyzing the first received ultrasonic wave, and may be determined by Equation 1 below.

[Equation 1]

[dB/cm]는 초음파 감쇠계수, d는 측정 대상체의 두께,

및

는 마주하는 한 쌍의 트랜스듀서 사이에 측정 대상체가 없는 경우와 있는 경우에 각각 수신된 신호의 파워스펙트럼레벨)(here,

[dB/cm] is the ultrasonic attenuation coefficient, d is the thickness of the measurement object,

and

is the power spectrum level of the received signal when there is no measurement object between a pair of transducers facing each other)

및

는 각각의 경우에 수신된 제1 수신초음파를 신호 처리(일 예로, 고속 푸리에 변환(FFT))하여 획득될 수 있다.At this time,

and

may be obtained by signal processing (eg, fast Fourier transform (FFT)) of the received first received ultrasound in each case.

And, in the operation step S310, the backscattering coefficient may be calculated by analyzing the first received ultrasound and the second ultrasound, and may be determined by Equation 2 below.

[Equation 2]

[cm^-1sr^-1]는 초음파 후방산란계수, R은 어느 하나의 트랜스듀서의 초음파 송수신면으로부터 측정 대상체의 중앙부까지의 거리,

와

는 각각 측정 대상체 및 측정 대상체의 중앙부에 대응하는 거리에 위치된 완전 반사체(스테인리스 스틸)로부터 후방 산란된 신호의 파워스펙트럼레벨, W(f)는 초음파 초점에서 -3 dB만큼의 빔폭(반치 빔폭; half-maximum beam width),

는 제2 수신초음파 중 측정 대상체 내부에서 후방 산란된 초음파 신호가 차지하는 시간, l은 측정 대상체의 중앙부로부터 표면까지의 거리,

및

는 각각 투과법에 의해 측정된 측정 대상체의 위상속도 및 감쇠계수) (here,

[cm ^-1 sr ^-1 ] is the ultrasonic backscattering coefficient, R is the distance from the ultrasonic transmission/reception surface of any one transducer to the center of the measurement object,

Wow

is the power spectrum level of the signal backscattered from the perfect reflector (stainless steel) located at a distance corresponding to the measurement object and the center of the measurement object, respectively, W(f) is the beam width by -3 dB at the ultrasonic focus (half full beam width; half-maximum beam width),

is the time occupied by the ultrasonic signal backscattered inside the measurement object among the second reception ultrasound waves , l is the distance from the center of the measurement object to the surface,

and

is the phase velocity and attenuation coefficient of the measurement object measured by the transmission method, respectively)

와

는 각각의 경우에 수신된 제2 수신초음파를 신호 처리(일예로, 고속 푸리에 변환(FFT))하여 획득할 수 있다. 또한, W(f)는 반치 빔폭(half-maximum beam width)으로, 예를 들어 구체적으로 설명하자면, 시간-진폭 그래프로 표현될 수 있는 제2 수신초음파를 신호 처리(일예로, 고속 푸리에 변환(FFT))하여, 결과의 일부가 포물선 형태의 그래프로 표현될 수 있는 주파수-진폭(log 스케일)로 변환하고, 획득된 신호 처리 결과에서, 최대 피크를 나타내는 주파수로부터 -3 dB되는 두 점으로부터 W(f)가 결정될 수 있다. At this time, in Equation 2

Wow

in each case may be obtained by signal processing (eg, fast Fourier transform (FFT)) of the received second ultrasonic wave. In addition, W (f) is a half-maximum beam width, for example, to describe in detail, the second received ultrasound signal processing (eg, fast Fourier transform ( FFT)), a part of the result is converted into a frequency-amplitude (log scale) that can be expressed as a graph in the form of a parabola, and in the obtained signal processing result, W from two points -3 dB from the frequency representing the maximum peak (f) can be determined.

는 제2 수신초음파 중 측정 대상체 내부에서 후방 산란된 초음파 신호가 차지하는 시간으로, 구체적으로 설명하면, 측정 대상체(M)로부터 반사된 초음파 신호 즉, 제2 수신초음파 중 측정 대상체(M)의 표면에서 반사된 초음파 신호를 제외하고, 측정 대상체(M) 내부에서 후방산란된 초음파 신호가 차지하는 시간이다.And,

is the time occupied by the ultrasonic signal backscattered inside the measurement object among the second ultrasonic reception waves. Specifically, the ultrasonic signal reflected from the measurement object M, that is, from the surface of the measurement object M among the second reception ultrasound waves It is the time occupied by the ultrasonic signal backscattered inside the measurement object M, excluding the reflected ultrasonic signal.

In addition, the attenuation coefficient of the measurement object M may be calculated by Equation 1 described above, and the phase velocity of the measurement object M may be calculated through the following reference equation.

[Reference formula]

[m/s]는 측정 대상체의 위상속도,

는 각진동수,

는 온도에 의존하는 수중에서의 음속(1482m/s, 20℃), d는 측정 대상체의 두께, Δ

는 초음파 전파 경로에 측정 대상체가 없는 경우와 있는 경우에 각각 수신된 신호의 위상차)(here,

[m/s] is the phase velocity of the measurement object,

is the angular frequency,

is the temperature-dependent speed of sound in water (1482 m/s, 20°C), d is the thickness of the object to be measured, Δ

is the phase difference of the received signal when there is no measurement object in the ultrasound propagation path, respectively)

If the attenuation coefficient and the backscattering coefficient of the measurement object M are calculated from the first receiving ultrasound and the second receiving ultrasound in the calculation step S310, the variable extraction step S320 calculates the frequency band calculated in the calculation step S310. Extract the attenuation coefficient and backscatter coefficient corresponding to a specific frequency (eg, 0.4 MHz) corresponding to the central band from the attenuation coefficient and backscatter coefficient corresponding to .

For reference, the ultrasonic attenuation coefficient and the backscattering coefficient for the frequency function calculated in the calculation step S310 are exemplified in FIGS. 4(a) and 4(b), respectively, and after the calculation is completed in the calculation step (S310) , in the variable extraction step ( S320 ), the attenuation coefficient and the backscattering coefficient corresponding to a specific frequency are extracted based on the result illustrated in FIG. 4 . And, the above-described extraction step (S300) may be performed in the extraction unit 120, and the operation step (S310) and the variable extraction step (S320) are performed in the operation part and the variable extraction part included in the extraction part 120, respectively. can be performed.

4. Prediction step <S400>

Prediction step (S400) is made in the prediction unit 140, any one of the bone density and bone structure of the measurement object (M) is a dependent variable, the extraction step (S300) to a multiple regression model consisting of at least two or more independent variables It is a step of predicting the bone density and bone structure of the measurement object (M) by applying the extracted attenuation coefficient and backscattering coefficient as independent variables. Here, the bone structure of the measurement object M includes at least one of a bone volume ratio (BT/TV), a trabecular trabecular thickness (Tb.Th), and a trabecular trabecular interval (Tb.Sp).

At this time, the multiple regression model used to predict the bone density and bone structure of the measurement object M in the prediction step (S400) is a multiple regression model performed before the first measurement step (S100) and the second measurement step (S200) It can be obtained through the creation step (not shown).

Here, in the multiple regression model generation step, the attenuation coefficient and backscattering coefficient measured in advance from a plurality of measurement object samples are used as independent variables, and any one of the bone density and bone structure of the measurement object sample is used as a dependent variable, and a plurality of independent In this step, multiple regression analysis is performed on the variable and each dependent variable, and a regression equation derived therefrom is obtained as a multiple regression model.

In this case, the multiple regression model can be obtained from Equation 3 below.

[Equation 3]

및

는 독립변수(초음파 감쇠계수 및 후방산란계수), a 및 b는 계수, c는 y절편) (where y is the dependent variable (bone density or bone structure),

and

is the independent variable (ultrasonic attenuation coefficient and backscattering coefficient), a and b are coefficients, c is the y-intercept)

To describe the multiple regression model in more detail, the multiple regression model expressed by Equation 3 may be obtained through multiple regression analysis on a plurality of pre-measured samples of the measurement object (M). Regression coefficients corresponding to a, b, and c can be derived from this, and bone density and bone structure can be predicted.

That is, the prediction unit 140 may pre-store as a multiple regression model a regression equation to which a regression coefficient derived from a pre-measured measurement object M sample is substituted. Specifically, the prediction unit 140 includes a multiple regression model for predicting bone density (BMD), a multiple regression model for predicting bone volume ratio (BV/TV), and a multiple regression model for predicting trabecular trabecular thickness (Tb.Th). , multiple regression models for trabecular spacing (Tb.Sp) prediction are stored, and thus, the prediction step (S400) applies the ultrasonic attenuation coefficient and the backscattering coefficient extracted in the extraction step (S300) to each multiple regression model. Bone density and bone structure can be evaluated together.

At this time, as shown in FIG. 5, the coefficient of determination of the multiple regression model (R ² , the square of the Pearson correlation coefficient R) has a larger value than the coefficient of determination of the linear regression model between a single independent variable and a dependent variable, When the multiple regression model is used, the bone density and bone structure of the measurement object M can be more accurately predicted. A detailed description thereof will be provided later.

4. Output step <S500>

The output step (S500) is a step of outputting the bone density and bone structure of the measurement object (M) predicted in the prediction step (S400). In this case, the predicted bone density and bone structure may be provided by the output unit 150 as at least one of an image, a sound, and an output. Here, the result output through the output unit 150 may include not only the bone density and bone structure of the measurement object M, but also osteoporosis diagnosis results that can be derived therefrom. That is, the osteoporosis diagnosis result obtained by evaluating the bone density and the bone structure together is output.

On the other hand, Figure 5 is an example to explain the reliability of the bone density and bone structure predicted from the multiple regression model to which the damping coefficient and the posterior coefficient are applied in predicting the bone density and bone structure of the measurement object, and the linear regression model and the multiple regression model This is a table comparing the coefficients of determination of the models.

In order to present the results of FIG. 5 , ultrasonic measurement was performed on a sample of the measurement object (M), and an ultrasonic attenuation coefficient and a backscattering coefficient for the sample of the measurement object (M) were extracted. At this time, the linear regression model is a regression analysis of each independent variable and the dependent variable by using any one of bone density and bone structure as a dependent variable, and attenuation coefficient or backscattering coefficient as an independent variable. Regression analysis of each dependent variable and independent variable was performed using any one of the dependent variables as the dependent variable and the attenuation coefficient and backscattering coefficient as independent variables.

Here, the sample to be measured (M) is 18 spongy bone samples having a rectangular parallelepiped shape prepared using the proximal part of a cow's femur having a structure and acoustic characteristics similar to those of the cancellous bone (T) of the human body, and a center frequency of 0.5 MHz The attenuation coefficient for the frequency region of 0.2 MHz to 0.6 MHz was calculated using the transmission method together with a pair of ultrasonic transducers 110 having A backscattering coefficient in the range of 0.2 MHz to 0.6 MHz was calculated using the pulse-echo method together with one ultrasonic transducer 110 having, and the backscattering coefficient corresponding to 0.4 MHz was extracted.

Then, by applying the extracted attenuation coefficient and backscattering coefficient to the linear regression model and the multiple regression model, the correlation between the dependent and independent variables of bone density and bone structure (BV/TV, Tb.Th, Tb.Sp), respectively. The coefficient of determination was obtained for .

Referring to FIG. 5 , it was found that the coefficient of determination of the multiple regression model had a larger value than that of the linear regression model. As described above, it can be confirmed that the bone density and bone structure of cancellous bone can be more accurately predicted using the multiple regression model using the two closely related ultrasound variables, the attenuation coefficient and the backscattering coefficient, as independent variables.

As described above, the specific description of the present invention has been made by the embodiments with reference to the accompanying drawings, but since the above-described embodiments have only been described with preferred examples of the present invention, the present invention is limited only to the above embodiments It should not be understood as being, and the scope of the present invention should be understood as the following claims and their equivalents.

100: Bone density and bone structure prediction device using ultrasonic damping coefficient and back scattering coefficient
110: transducer
120: measurement unit
130: extraction unit
140: prediction unit
150: output unit
M: object to be measured
C: cortical bone
T: spongy bone
Tb: bone shochu
Tb.Sp : Trabecular distance
Tb.Th: bone trabecular thickness

Claims (8)

As a method for predicting bone density and bone structure using ultrasonic damping coefficient and backscattering coefficient,
A first measurement step of irradiating a transmitting ultrasound wave to a measurement object by any one transducer, and receiving a first receiving ultrasound wave transmitted through the measurement object by another transducer facing the one transducer;
a second measurement step of irradiating a transmission ultrasound wave toward the measurement object by any one transducer, and receiving a second reception ultrasound wave returned by reflecting the measurement object;
an extraction step of extracting an attenuation coefficient and a backscattering coefficient for a frequency function by analyzing the electrical signals of the first ultrasonic wave and the second ultrasonic wave received; and
Any one of the bone density and bone structure of the measurement object is a dependent variable, and the attenuation coefficient and the backscattering coefficient extracted in the extraction step are applied as independent variables to a multiple regression model in which at least two or more variables are independent variables. Predicting the bone density and bone structure of the prediction step; including,
The extraction step is
Calculation step of calculating an attenuation coefficient and a backscattering coefficient for a frequency function in a frequency band of 0.2 MHz to 0.6 MHz (hereinafter referred to as a 'calculation frequency band') through the electrical signals of the first and second reception ultrasound waves ; and
variable extraction step of extracting an attenuation coefficient and a backscattering coefficient corresponding to the center band from the calculated frequency band;
In the calculation step, the attenuation coefficient may be calculated by analyzing the first received ultrasonic wave, and is determined by Equation 1 below,
[Equation 1]

- here,

[dB/cm] is the ultrasonic attenuation coefficient, d is the thickness of the measurement object,

and

is the power spectrum level of the received signal when there is no measurement object between a pair of transducers facing each other ―
In the calculation step, the backscattering coefficient may be calculated by analyzing the first received ultrasound and the second received ultrasound, and is determined by the following Equation 2,
[Equation 2]

- here,

[cm ^-1 sr ^-1 ] is the ultrasonic backscattering coefficient, R is the distance from the ultrasonic transmission/reception surface of any one transducer to the center of the measurement object,

Wow

is the power spectrum level of the signal backscattered from the perfect reflector (stainless steel) located at a distance corresponding to the measurement object and the center of the measurement object, respectively, W(f) is the beam width by -3 dB at the ultrasonic focus (half full beam width; half-maximum beam width),

is the time occupied by the ultrasonic signal backscattered inside the measurement object among the second reception ultrasound waves , l is the distance from the center of the measurement object to the surface,

and

are the phase velocity and attenuation coefficient of the measurement object measured by the transmission method, respectively.
Bone density and bone structure prediction method using ultrasonic damping coefficient and backscattering coefficient.
delete delete delete According to claim 1,
The bone structure of the measurement target is characterized in that it includes at least one of a bone volume ratio (BT/TV), a trabecular trabecular thickness (Tb.Th), and a trabecular trabecular interval (Tb.Sp).
Bone density and bone structure prediction method using ultrasonic damping coefficient and backscattering coefficient.
6. The method of claim 5,
The attenuation coefficient and backscattering coefficient measured in advance from a plurality of measurement object samples are used as independent variables, and any one of the bone density and bone structure of the measurement object sample is used as a dependent variable, a multiple regression model generation step of performing multiple regression analysis and obtaining an index equation derived therefrom as the multiple regression model; characterized in that it comprises
Bone density and bone structure prediction method using ultrasonic damping coefficient and backscattering coefficient.
7. The method of claim 6,
In the multiple regression model generation step, the multiple regression model is obtained from Equation 3 below,
[Equation 3]

― where y is the dependent variable (bone density or bone structure),

and

is the independent variable (attenuation coefficient and backscatter coefficient), a and b are the coefficients, and c is the y-intercept ―
The coefficient of determination of the multiple regression model (R ² , the square of the Pearson correlation coefficient R) is characterized in that it has a larger value than the coefficient of determination of the linear regression model between a single independent variable and a dependent variable.
Bone density and bone structure prediction method using ultrasonic damping coefficient and backscattering coefficient.
According to claim 1,
an output step of outputting the bone density and bone structure of the measurement target predicted in the prediction step; characterized in that it further comprises
Bone density and bone structure prediction method using ultrasonic damping coefficient and backscattering coefficient.

Images