Diagnostic analysis of ultrasound data
A method and apparatus are provided for investigating tissue in which acoustic data are derived from scattering a plurality of pulsed spherical or cylindrical acoustic waves from a plurality of transmission elements through the tissue to a plurality of receiving elements. The acoustic data, which include a mix of reflected and transmitted acoustic waves, are received and digitized, and a representation of a portion of the tissue is generated from the digitized acoustic data.
1. A method for investigating tissue, the method comprising:
receiving acoustic data derived from scattering a plurality of pulsed spherical or cylindrical acoustic waves from a plurality of transmission elements through the tissue to a plurality of receiving elements, wherein the received acoustic data include a mix of reflected and transmitted acoustic waves;
digitizing the acoustic data; and
generating a representation of a portion of the tissue from the digitized acoustic data, wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises removing a direct coupling pulse from the received acoustic data.
2. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises performing a time-series analysis of the digitized acoustic data.
3. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises performing a frequency-series analysis of the digitized acoustic data.
4. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises:
for each pair of transmitting and receiving elements, computing a wave path from a time delay; and
summing such wave paths to derive an image representation of the portion of the tissue.
5. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises:
organizing the digitized acoustic data according to at least two independent dimensions; and
numerically fitting the digitized data to an expression relating the expected intensity variation in terms of physical quantities having different dependencies in the at least two independent dimensions.
6. The method recited in claim 5 wherein the physical quantities comprise a compressibility contrast and a density contrast.
7. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises:
converting the digitized acoustic data to a frequency domain;
extracting scattered-field Fourier components from the converted data;
calculating a complex potential from a scattered field defined by the scattered-field Fourier components; and
deriving physical properties of the portion of the tissue from the complex potential.
8. The method recited in claim 7 wherein the physical properties comprise a sound speed and an attenuation.
9. The method recited in claim 1 wherein generating the representation of the portion of the tissue from the digitized acoustic data comprises:
(a) initially choosing a tissue model of the portion of the tissue;
(b) simulating wave propagation of the pulsed spherical or cylindrical waves through the tissue model with a wave-propagation model to determine simulated data;
(c) backpropagating a residual derived from comparing the simulated data with the digitized acoustic data through an adjoint of the wave-propagation model to update the tissue model; and
(d) repeating (b) and (c) until a magnitude of the residual is less than a predefined threshold.
10. The method recited in claim 1 wherein:
the acoustic data comprise a first set of acoustic data measured from a first insonification of the tissue and a second set of acoustic data measured from a second insonification of the tissue; and
generating the representation comprises determining a compressibility of the tissue from the first and second sets of acoustic data.
11. The method recited in claim 10 wherein measurements of the first and second sets of acoustic data are separated by a mechanical compression of the tissue.
12. The method recited in claim 10 wherein the second insonification causes a compression of the tissue.
13. The method recited in claim 1 wherein:
the acoustic data comprise a first set of acoustic data measured from a first insonification of the tissue under a first compression mode and a second set of acoustic data measured from a second insonification of the tissue under a second compression mode; and
generating the representation comprises determining an acoustic property of the tissue correlated with a difference between the first and second compression modes.
14. The method recited in claim 13 wherein the acoustic property comprises a property selected from the group consisting of sound speed, attenuation, density, compressibility, absorption, acoustic impedance change, and blood flow rate.
15. A method for investigating tissue, the method comprising:
receiving acoustic data derived from scattering a radiation pattern of acoustic waves from a plurality of transmission elements to a plurality of receiving elements, wherein the received acoustic data include a mix of reflected and transmitted acoustic waves;
digitizing the acoustic data;
organizing the digitized data according to at least two independent dimensions; and
numerically fitting the digitized data to an expression relating the expected intensity variation in terms of physical quantities having different dependencies in the at least two independent dimensions.
16. The method recited in claim 15 wherein the physical quantities comprise a compressibility contrast and a density contrast.
17. The method recited in claim 16 wherein the at least two dimensions comprise a radial dimension and an angular dimension, and wherein the compressibility contrast does not vary with the angular dimension.
18. The method recited in claim 17 wherein the expression comprises
P
s
=
P
0
(
κ
s
-
κ
0
κ
0
+
3
ρ
s
-
ρ
0
2
ρ
s
+
ρ
0
cos
θ
)
ⅇ
ik
0
r
r
,
wherein r is the radial dimension, θ is the angular dimension, κ s is a compressibility of a feature within the tissue, ρ s is a density of the feature, P s is an intensity of the feature within the tissue, κ 0 is a compressibility of a region surrounding the feature, ρ 0 is a density of the region surrounding the feature, k 0 is a wave number of the region surrounding the feature, and P 0 is an intensity of the region surrounding the feature.
19. A method for investigating tissue, the method comprising:
receiving scattered acoustic data derived from scattering an initial acoustic radiation pattern from the tissue, wherein the received scattered acoustic data include a mix of reflected and transmitted acoustic waves;
digitizing the received acoustic data and the initial acoustic radiation pattern;
converting the digitized received acoustic data and the digitized initial acoustic radiation pattern to a frequency domain;
extracting scattered-field components by comparing the converted digitized received acoustic data with the converted initial acoustic radiation pattern;
calculating a complex potential from the scattered-field components; and
deriving physical properties of a portion of the tissue from the complex potential.
20. The method recited in claim 19 wherein calculating the complex potential comprises solving a scattering equation within a Born approximation.
21. The method recited in claim 19 wherein calculating the complex potential comprises solving a scattering equation within a Rytov approximation.
22. The method recited in claim 19 wherein calculating the complex potential comprises unwrapping a phase.
23. The method recited in claim 22 wherein unwrapping the phase comprises:
fitting an Nth-order polynomial through N+1 unwrapped phase values;
extrapolating the polynomial to a subsequent unwrapped phase value to determine an extrapolated value;
calculating a difference between the extrapolated value and the subsequent unwrapped phase value; and
adding a rounded multiple of 2π to the subsequent unwrapped phase value.
24. The method recited in claim 22 wherein unwrapping the phase comprises:
fitting a curve with M unwrapped phase values;
extrapolating the curve to a subsequent unwrapped phase value to determine an extrapolated value;
calculating a difference between the extrapolated value and the subsequent unwrapped phase value; and
adding a rounded multiple of 2π to the subsequent unwrapped phase value.
25. The method recited in claim 24 wherein the curve comprises an Nth order polynomial with N<M−1.
26. The method recited in claim 24 wherein the curve comprises a spline.
27. A method for investigating tissue, the method comprising:
(a) initially choosing a tissue model for a portion of the tissue;
(b) initially choosing a source position;
(c) simulating, for the source position, propagation of an acoustic radiation pattern through the tissue model with a wave-propagation model to determine simulated data;
(d) backpropagating a residual derived from comparing the simulated data with a set of measured data derived by propagating the acoustic radiation pattern through the portion of the tissue physically through an adjoint of the wave-propagation model to update the tissue model, wherein the set of measured data includes a mix of reflected and transmitted acoustic waves;
(e) selecting a new source position; and
(f) repeating (c)-(e) until a magnitude of the residual is less than a predefined threshold.
28. The method recited in claim 27 wherein initially choosing a tissue model comprises performing a full-aperture tomographic analysis of the set of measured data.
29. The method recited in claim 27 wherein initially choosing a tissue model comprises performing a diffraction tomographic analysis of the set of measured data.
30. A method for unwrapping a phase of a function acting on a complex quantity, the method comprising:
fitting an Nth-order polynomial through N+1 unwrapped phase values;
extrapolating the polynomial to a subsequent unwrapped phase value to determine an extrapolated value;
calculating a difference between the extrapolated value and the subsequent unwrapped phase value; and
adding a rounded multiple of 2π to the subsequent unwrapped phase value.
31. A method for unwrapping a phase of a function acting on a complex quantity, the method comprising:
fitting a curve with M unwrapped phase values;
extrapolating the curve to a subsequent unwrapped phase value to determine an extrapolated value;
calculating a difference between the extrapolated value and the subsequent unwrapped phase value; and
adding a multiple of 2π to the unwrapped phase value.
32. The method recited in claim 31 wherein the curve comprises an Nth-order polynomial with N<M−1.