Ultrasonic bone assessment method and apparatus

• US 5,651,363 A
• Filed: 02/16/1996
• Issued: 07/29/1997
• Est. Priority Date: 02/16/1996
• Status: Expired due to Fees

• Current Assignee: DJO LLC (Enovis Corp.)
• Original Assignee: OrthoLogic Corporation
• Inventors: Kaufman, Jonathan J., Chiabrera, Alessandro
• Primary Examiner(s): Lateef, Marvin M.
• Assistant Examiner(s): FIELDS, DERRICK LAMAR
• Application Number:
  US08/602,410
• Time in Patent Office: 529 Days
• Field of Search: 128/660.02, 128/660.01, 601/2
• US Class Current: 600/438
• CPC Class Codes: A61B 8/0875 for diagnosis of bone A61B5/4504 takes precedence

Abstract

A method for assessment of bone properties comprises passing ultrasonic signal through a bone tissue, modeling an output signal and a reference output signal and subjecting parameters of the models to comparative analysis. An apparatus is also disclosed for practicing the method.

Claims

1. A method of non-invasive and quantitative assessment of the status of bone tissue in vivo for one or more of the quantities:

bone-mineral density, strength, and fracture risk, comprising the steps of;

(a) acoustically coupling a pair of transducers to nearby skin on opposite sides of said bone tissue;

(b) generating an ultrasound excitation signal and directing same from one transducer to another transducer of said pair of transducers through said bone tissue, thereby producing a bone-oriented electrical output signal of the form of z_s (t)=p_s (t)+n_s (t), where p_s (t) is said bone-oriented output signal per se and n_s (t) is an additive, uncorrelated Gaussian measurement noise associated with said bone-oriented signal, said excitation signal being a finite-duration signal repeated substantially in a range from 1 to 1000 Hz and consisting of plural frequencies spaced in an ultrasonic spectral region up to about 2 MHz;

(c) independently directing said excitation signal from said one transducer to said another transducer through a medium with known acoustic properties and path length and free of said bone tissue, thereby producing a reference electrical output signal of the form of z_r (t)=p_r (t)+n_r (t), where p_r (t) is said reference output signal per se and n_r (t) is an additive, uncorrelated Gaussian measurement noise associated with said reference signal;

(d) parametric modeling said bone-oriented signal and said reference signal, with obtaining two parametric models thereof, p_s (t) and p_t (t), respectively, to thereby establish a set Θ

_s of bone-oriented parameters and a set Θ

_r of reference parameters correspondingly associated with said models, and

(e) subjecting said two sets of parameters to comparative analysis, whereby an estimate of said one or more quantities is obtained.

2. The method according to claim 1, wherein said parametric model of said bone-oriented output signal has the form of

3. The method according to claim 1, wherein said parametric model of said bone-oriented output signal has the form of

4. The method according to claim 2, wherein said parametric model of said bone-oriented output signal has the form of

5. The method according to claim 3, wherein said parametric model of said bone-oriented output signal has the form of

6. The method according to claim 1, wherein said parametric model of said reference signal has the form of

7. The method according to claim 1, wherein said parametric model of said reference signal has the form of

8. The method according to claim 6, wherein said parametric model of said reference signal has the form of

9. The method according to claim 7, wherein said parametric model of said reference signal has the form of

10. The method according to claim 1, wherein said set Θ

_s of bone-oriented parameters is estimated with the use of a least square optimization algorithm, to thereby provide a maximum likelihood estimate Θ

_s,ML defined as ##EQU5## where argmin denotes a value of said bone-oriented parameters which provides a smallest value of said sum Σ

.

11. The method according to claim 1, wherein said set Θ

_r of reference parameters is estimated with the use of a least square optimization algorithm, to thereby provide a maximum likelihood estimate Θ

_r,ML defined as ##EQU6## where argmin denotes a value of said reference parameters which provides a smallest value of said sum Σ

.

12. The method according to claim 1, wherein said step of comparative analysis is performed with the use of one or more of a plurality of associated parameters:

age, bony member thickness, sex, height, weight specific for an individual patient.

13. The method according to claim 1, further comprising a preprocessing step prior to said step of comparative analysis, inputs of said preprocessing step being said two sets of parameters of said models and an output of said preprocessing step being a reduced set of phase responsive parameters characterizing said models and associated with said bone tissue.

14. The method according to claim 13, further comprising one or more of a plurality of associated parameters:

age, bone tissue thickness, sex, height, weight, specific for an individual patient, input to said preprocessing step.

15. The method according to claim 13, wherein said preprocessing step further includes determination of an optimal transfer function estimate, said estimate being said output of said preprocessing step.

16. The method according to claim 1, wherein said comparative analysis is performed with the use of a neural network configured to generate an estimate of said one or more of said quantities from said sets of parameters and from said associated parameters.

17. The method according to claim 1, wherein said comparative analysis is performed with the use of multivariate regressions.

18. The method according to claim 1, wherein said comparative analysis is performed with the use of a statistical hypothesis testing algorithm.

19. A method of non-invasive and quantitative assessment of the status of bone tissue in vivo for one or more of the quantities:

bone-mineral density, strength, and fracture risk, comprising the steps of;

(a) acoustically coupling a pair of transducers to nearby skin on opposite sides of said bone tissue;

(b) generating an ultrasound excitation signal and directing same from one transducer to another transducer of said pair of transducers through said bone tissue, thereby producing a bone-oriented electrical output signal of the form of z_s (t)=p_s (t)+n_s (t), where p_s (t) is said bone-oriented output signal per se and n_s (t) is an additive, uncorrelated Gaussian measurement noise associated with said boneoriented signal, said excitation signal being a finite-duration signal repeated substantially in a range from 1 to 1000 Hz and consisting of plural frequencies spaced in an ultrasonic spectral region up to about 2 MHz;

(c) independently directing said excitation signal from said one transducer to said another transducer through a medium with known acoustic properties and path length and free of said bone tissue, thereby producing a reference electrical output signal of the form of z_r (t)=p_r (t)+n_r (t), where p_r (t) is said reference output signal per se and n_r (t) is an additive, uncorrelated Gaussian measurement noise associated with said reference signal;

(d) parametric modeling said bone-oriented signal and said reference signal, with obtaining two parametric models thereof, p_s (t) and p_r (t), respectively, to thereby establish a set Θ

_s of bone-oriented parameters and a set Θ

_r of reference parameters correspondingly associated with said models;

(e) preprocessing said two sets of parameters of said models, an output of said preprocessing being a reduced set of parameters characterizing said models; and

(f) subjecting said two sets of parameters to comparative analysis, whereby an estimate of said one or more quantities is obtained.

20. A method of non-invasive and quantitative assessment of the status of bone tissue in vivo for at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality comprising the steps of:

(a) acoustically coupling a pair of transducers to nearby skin on opposite sides of said bone tissue;

(b) generating an ultrasound signal and directing said ultrasound signal from one transducer to another transducer of said pair of transducers through said bone tissue, to produce a bone-oriented electrical output signal z_s (t), said ultrasound signal being a finite-duration signal repeated substantially in a range of from about 1 to about 1000 Hz and having plural frequencies spaced in an ultrasonic spectral region up to about 2 MHz;

(c) independently directing said ultrasound signal from said one transducer to said another transducer through a medium with known acoustic properties and path length and free of said bone tissue to produce a reference electrical output signal z_r (t);

(d) establishing a set Θ

_s of bone-oriented parameters associated with said bone-oriented output signal z_s (t) and a set Θ

_r of reference parameters associated with said reference signal z_r (t); and

(e) subjecting said set Θ

_s of bone-oriented parameters and said set Θ

_r of reference parameters to comparative analysis, whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

21. The method according to claim 20, wherein said set Θ

_s of bone-oriented parameters includes a mean instantaneous frequency associated with said bone-oriented output signal z_s (t) and said set Θ

_r of reference parameters includes a mean instantaneous frequency associated with said reference signal z.sub. (t).

22. The method according to claim 21, wherein step (d) includes the steps of:

processing said bone-oriented output signal z_s (t) to obtain a corresponding Hilbert transform, z_sH (t);

obtaining a complex analytic signal z_sA (t) associated with said bone-oriented output signa z_s (t) wherein said complex analytic signal z_sA (t) is defined as z_sA (t)=z_s (t)+jz_sH (t), where j=(-1)^1/2 ;

processing said complex analytic signal z_sA (t) to obtain said mean instantaneous frequency associated with said bone-oriented output signal z_s (t);

processing said reference signal z_r (t) to obtain a corresponding Hilbert transform, z_rH (t);

obtaining a complex analytic signal z_rA (t) associated with said reference signal z_r (t) wherein said complex analytic signal Z_rA (t) is defined as Z_rA (t)=Z_r (t)+jz_rH (t), where j=(-1)^1/2 ; and

processing said complex analytic signal z_rA (t) to obtain said mean instantaneous frequency associated with said reference signal z_r (t).

23. The method according to claim 21, wherein step (e) includes the step of evaluating a quantitative ultrasonic feature parameter β

wherein β

=1-(MIF/MIF_ref) where MIF is said mean instantaneous frequency associated with said bone-oriented output signal z_s (t) and MIF_ref is said mean instantaneous frequency associated with said reference signal z_r (t), to thereby obtain an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture, and bone quality.

24. The method according to claim 20, wherein said set Θ

_s of bone-oriented parameters includes an average frequency content of said bone-oriented output signal z_s (t) and said set Θ

_r of reference parameters includes an average frequency content of said reference signal z_r (t).

25. The method according to claim 24, wherein step (d) includes the steps of:

using zero-crossing analysis to define said average frequency content of said bone-oriented output signal z_s (t); and

using zero-crossing analysis to define said average frequency content of said reference signal z_r (t).

26. The method according to claim 20, wherein said set Θ

_s of bone-oriented parameters includes an envelope time duration parameter of said bone-oriented output signal z_s (t) and said set Θ

_r of reference parameters includes an envelope time duration parameter of said reference signal z_r (t).

27. The method according to claim 26, wherein step (d) includes the steps of:

processing said bone-oriented output signal z_s (t) to obtain a corresponding Hilbert transform z_sH (t);

obtaining a complex analytic signal z_sA (t) associated with said bone-oriented output signal z_s (t) wherein said complex analytic signal z_sA (t) is defined as Z_sA (t)=Z_s (t)+jz_sH (t), where j=(-1)^1/2 ;

processing said complex analytic signal z_sA (t) to obtain said envelope time duration parameter of said boneoriented output signal z_s (t);

processing said reference signal z_r (t) to obtain a corresponding Hilbert transform z_rH (t);

obtaining a complex analytic signal z_rA (t) associated with said reference signal z_r (t) wherein said complex analytic signal z_rA (t) is defined as z_rA (t)=z_r (t)+jz_rH (t), where j=(-1)^1/2 ; and

processing said complex analytic signal z_rA (t) to obtain said envelope time duration parameter of said reference signal z_r (t).

28. The method according to claim 26, wherein step (d) includes the steps of:

using an envelope detection method to obtain said envelope time duration parameter of said bone-oriented output signal z_s (t); and

using an envelope detection method to obtain said envelope time duration parameter of said reference signal z_r (t).

29. The method according to claim 20, wherein said set Θ

_s of bone-oriented parameters includes an overall energy content of said bone-oriented output signal z_s (t) and said set Θ

_r of reference parameters includes an overall energy content of said reference signal z_r (t).

30. The method according to claim 20, wherein said set Θ

_s of bone-oriented parameters includes an envelope time delay parameter of said bone-oriented output signal z_s (t) and said set Θ

_r of reference parameters includes an envelope time delay parameter of said reference signal z_r (t).

31. The method according to claim 20 further comprising a first step, prior to step (e), of obtaining a differential phase estimate τ

_g (f) associated with the phase difference between said bone-oriented output signal z_s (t) and said reference signal z_r (t); and

wherein step (e) includes subjecting said set Θ

_s of bone-oriented parameters, said set Θ

_r of reference parameters and said differential phase estimate τ

_g (f) to comparative analysis, whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

32. The method according to claim 31, wherein said first step prior to step (e) includes:

processing said bone-oriented output signal z_s (t) to obtain a corresponding Fourier transform Z_s (f);

processing said reference signal z_r (t) to obtain a corresponding Fourier transform Z_r (f); and

generating said differential phase estimate τ

_g (f) wherein τ

_g (f)=(-L/2II)*(dβ

(f)/df), where β

(f)=arg[H(f)]/L where H(f)=Z_s (f)/Z_r (f) and L is a thickness of a bony member.

33. The method according to claim 32, wherein L is a thickness of said bone tissue.

34. The method according to claim 31, further comprising a second step, prior to step (e), of processing said differential phase estimate τ

_g (f) to obtain a group velocity estimate v_g (f); and

wherein step (e) includes subjecting said set Θ

_s of bone-oriented parameters, said set Θ

_r of reference parameters, and said group velocity estimate v_g (f) to comparative analysis, whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

35. The method according to claim 34, further comprising a third step, prior to step (e), of processing said group velocity estimate v_g (f) to obtain a set of polynomial coefficients;

and

wherein step (e) includes subjecting said set Θ

_s of bone-oriented parameters, said set Θ

_r of reference parameters, and said set of polynomial coefficients to comparative analysis, whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

36. The method according to claim 20 wherein step (e) includes subjecting said set Θ

_s of bone-oriented parameters, said set Θ

_r of reference parameters and at least one value selected from the group consisting of a patient'"'"'s age, a patient'"'"'s sex, a patient'"'"'s height, a patient'"'"'s weight, a thickness of said bone tissue, and a thickness of a bony member to comparative analysis, whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

37. The method according to claim 20, wherein said comparative analysis is performed with the use of a neural network.

38. The method according to claim 20, wherein said comparative analysis is performed with the use of multivariate regressions.

39. The method according to claim 20, wherein said comparative analysis is performed with the use of a statistical hypothesis testing algorithm.

40. An apparatus for non-invasive and quantitative assessment of the status of bone tissue in vivo for at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture, and bone quality comprising

a pair of ultrasonic transducers;

means for generating an ultrasound signal, said ultrasound signal being a finite-duration signal repeated substantially in a range of from about 1 to about 1000 Hz and having plural frequencies spaced in an ultrasonic spectral region up to about 2 MHz;

means for directing said ultrasound signal from one transducer to another transducer of said pair of transducers through said bone tissue to produce a bone-oriented electrical output signal z_s (t);

means for independently directing said ultrasound signal from said one transducer to said another transducer of said pair of transducers through a medium of known acoustic properties and path length and free of said bone tissue to produce a reference electrical output signal z_r (t);

means for establishing a set Θ

_s of bone-oriented parameters associated with said bone-oriented output signal z_s (t) and a set Θ

_r of reference parameters associated with said reference signal z_r (t); and

means for performing comparative analysis of said set of bone-oriented parameters Θ

_s and said set of reference parameters Θ

_r to thereby obtain an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality.

41. The apparatus of claim 40, further comprising means for preprocessing said set of bone-oriented parameters Θ

_s and said set of reference parameters Θ

_r prior to performing said comparative analysis in order to obtain a reduced set of bone-oriented parameters and a reduced set of reference parameters; and

wherein said performing means performs said comparative analysis of said reduced set of bone-oriented parameters and said reduced set of reference parameters to thereby obtain an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality.

42. The apparatus of claim 40, further comprising:

means for averaging a plurality of bone-oriented output signals z_s (t) to produce an averaged bone-oriented output signal;

means for averaging a plurality of reference signals z_r (t)to produce an averaged reference signal; and

wherein said establishing means produces a set Θ

_s of bone-oriented parameters associated with said averaged bone-oriented output signal and a set Θ

_r of reference parameters associated with said averaged reference signal.

43. A method of non-invasive and quantitative assessment of the status of bone tissue in vivo for at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality comprising the steps of:

(a) acoustically coupling a pair of transducers to nearby skin on opposite sides of said bone tissue;

(b) generating an ultrasound signal and directing said ultrasound signal from one transducer to another transducer of said pair of transducers through said bone tissue, to produce a bone-oriented electrical output signal, said ultrasound signal being a finite-duration signal having plural frequencies spaced in an ultrasonic spectral region up to about 2 MHz;

(c) repeating step (b) a plurality of times to obtain a plurality of bone-oriented output signals;

(d) averaging said plurality of bone-oriented output signals to obtain an averaged bone-oriented output signal;

(e) independently directing said ultrasound signal from said one transducer to said another transducer through a medium with known acoustic properties and path length and free of said bone tissue to produce a reference electrical output signal;

(f) repeating step (e) a plurality of times to obtain a plurality of reference signals;

(g) averaging said plurality of reference signals to obtain an averaged reference signal;

(h) establishing a set Θ

_s of bone-oriented parameters associated with said averaged bone-oriented output signal and a set Θ

_r of reference parameters associated with said averaged reference signal; and

(i) subjecting said set Θ

_s of bone-oriented parameters and said set Θ

_r of reference parameters to comparative analysis whereby an estimate of said at least one of the quantities, bone-mineral density, bone strength, bone fracture risk, bone architecture and bone quality is obtained.

44. The method according to claim 43, wherein said comparative analysis is performed with the use of a neural network.

45. The method according to claim 43, wherein said comparative analysis is performed with the use of multivariate regressions.

46. The method according to claim 43, wherein said comparative analysis is performed with the use of a statistical hypothesis testing algorithm.