Method for enhancement of ultrasonic image data

• US 4,858,124 A
• Filed: 08/11/1987
• Issued: 08/15/1989
• Est. Priority Date: 08/15/1984
• Status: Expired due to Term

• Current Assignee: Riverside Research Institute
• Original Assignee: Riverside Research Institute
• Inventors: Yaremko, Mykola M., Silverman, Ronald H., Lizzi, Frederic L., Coleman, D. Jackson
• Primary Examiner(s): Goldberg, E. A.
• Assistant Examiner(s): Donovan, Lincoln
• Application Number:
  US07/085,289
• Time in Patent Office: 735 Days
• Field of Search: 364/414.5, 128/660, 128/653, 73/602
• US Class Current: 600/443
• CPC Class Codes: A61B 8/10 Eye inspection
• CPC Class Codes: A61B 8/5207 involving processing of raw data to produce diagnostic data, e.g. for generating an image reconstruction G06T11/00
• CPC Class Codes: G01N 29/0645 Display representation or displayed parameters, e.g. A-, B- or C-Scan
• CPC Class Codes: G01N 29/348 with frequency characteristics, e.g. single frequency signals, chirp signals measuring frequency of mechanical vibrations or acoustic waves in general G01H1/06, G01H3/04; measuring frequency or analysing frequency spectra G01R23/00
• CPC Class Codes: G01S 7/52036 using analysis of echo signal for target characterisation
• CPC Class Codes: G01S 7/52073 Production of cursor lines, markers or indicia by electronic means
• CPC Class Codes: G01S 7/52074 Composite displays, e.g. split-screen displays; Combination of multiple images or of images and alphanumeric tabular information

Abstract

Methods for enhancement of data obtained from scanning objects such as body tissue with ultrasonic wideband pulses to detect characteristics of the tissue and to identify the tissue. The methods include performing spectral analysis on selected time portions of signals obtained from ultrasonic scanning to derive spectral data which is representative of the received signal characteristics for spatial samples of a particular region of interest in the body. The characteristic values are selected to correlate with significant characteristics of the material. A display of the selected body region is generated, with the display characteristic for each of the spatial samples being selected in accordance with the derived spectral characteristic values. Examples of the spectral characteristic values include spectral amplitude, spectral slope, and spectral amplitude uncertainty. The spectral characteristic data can be used to derive periodicity data, such as cepstral data or spatial-correlation data, for each of the spatial samples. The spectral characteristic data obtained can be compared to data for tissue having known tissue characteristics to identify the tissue type. Related means for suppressing the display of noise, characterizing tissues using discriminent functions based on spectral data, and estimating scatter sizes are also shown.

Claims

1. A method for processing reflected acoustic signals from a body of material and for generating a display of at least a portion of said body, representative of characteristics of said material comprising:

transmitting acoustic signals into a region of interest in said body in an ordered spatial progression;

receiving reflected acoustic signals from said body;

analyzing the frequency content of selected time portions of said received signal to derive spectral data representative of spectral power as a function of frequency for spatial samples of said body within said body region of interest;

said spatial samples corresponding to said time portions of said received signal;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

deriving spectral characteristic values from said normalized spectral data independently for each of said spatial samples, said spectral characteristic values being selected from the group comprising frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty; and

generating a display of said body region of interest with a display characteristic for each of said spatial samples selected in accordance with said derived spectral characteristic values.

2. A method as specified in claim 1 wherein said step of deriving spectral characteristic values comprises deriving a value representative of frequency specific normalized spectral amplitude for each of said samples.

3. A method as specified in claim 1 wherein said step of deriving spectral characteristic values comprises deriving a value representative of spectral slope for each of said samples.

4. A method as specified in claim 1 wherein said step of deriving spectral characteristic values comprises deriving a value representative of spectral amplitude uncertainty for each of said samples.

5. A method as specified in claim 1 wherein said step of deriving spectral characteristic values comprise deriving at least two of said spectral characteristic values;

and deriving combined spectral characteristic values from said at least two sets of spectral characteristic values, said combined spectral characteristic values being derived to correlate with significant material characteristics; and

wherein said step of generating a display comprises generating a display characteristic for each spatial sample in accordance with said derived combined spectral characteristic values.

6. A method as specified in claim 5 wherein said step of deriving spectral characteristic values comprises deriving at least three sets of said spectral characteristic values and deriving said combined spectral characteristic values from said three sets.

7. A method as specified in claim 1 wherein said step of generating a display comprises generating a display for each spatial region with a color selected in accordance with said derived spectral characteristic value.

41. A method as specified in claim 1, 8, 9, 10, 17, 18, 22, 25, or 32 wherein the steps of said method are repeated at least two different times, and wherein the results of each final step are compared to detect any changes in the characteristics of said body.

8. A method for processing reflected acoustic signals from a body of material and for generating a display of at least a portion of a cross-section of said body, representative of characteristics of said material, comprising:

transmitting acoustic signals into a region of interest in said body in an ordered spatial progression corresponding to said cross-section;

receiving reflected acoustic signals from said body, said signals comprising signal time intervals corresponding to reflections from successive linear acoustic paths through said material in said spatial progression;

dividing selected time periods of said received signals into signal time portions, each signal time portion comprising a selected time sample of one of said signal time intervals and corresponding to a spatial sample of said material along one of said acoustic paths in said region of interest;

performing a spectral analysis of each of said signal time portions to derive therefrom spectral data representative of the spectral power level of said signal time portion at various frequencies within the frequency band of said acoustic pulse signals;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

deriving at least one spectral characteristic value from said normalized spectral data independently for each of said signal time portions corresponding to a segment of an acoustic path in a region of interest of said cross-section, said spectral characteristic value being selected from the group consisting of frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainity;

and generating a display of said region of interest of said cross-section, said display having elemental display areas corresponding to said spatial samples, each of said display areas having a display characteristic representative of said spectral characteristic value.

9. A method for processing reflected acoustic signals from body tissue and for generating a display of at least a portion of said tissue, representative of clinicaly significant characteristics of said tissue comprising:

transmitting ultrasonic signals into a region of interest of said tissue in an ordered spatial progression;

receiving reflected ultrasonic signals from said tissue;

performing a spectral analysis of selected time portions of said received signal to derive spectral data representative of spectral power as a function of frequency for spatial samples of said tissue within said region of interest, said spatial samples corresponding to said time portions of said received signal;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

deriving spectral characteristic values from said normalized spectral data independently for each of said spatial samples, said spectral characteristic values being selected from the group comprising frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty; and

generating a display of said tissue regionof interest with a display characteristic for each of said spatial samples selected in accordance with said derived spectral characteristic values.

10. A method as specified in claim 9 wherein said step of deriving spectral characteristic values comprises deriving a value representative of frequency specific normalized spectral amplitude for each of said samples.

11. A method as specified in claim 9 wherein said step of deriving spectral characteristic values comprises deriving a value representative of spectral slope for each of said samples.

12. A method as specified in claim 9 wherein said step of deriving spectral characteristic values comprises deriving a value representative of spectral amplitude uncertainty for each of said samples.

13. A method as specified in claim 9 wherein said step of deriving spectral characteristic values comprises deriving at least two spectral characteristic values selected from said group;

and deriving combined spectral characteristic values from said at least two sets of spectral characteristic values, said combined spectral characteristic values being derived to correlate with clinically significant tissue characteristics; and

wherein said step of generating a display comprises generating a display characteristic for each spatial sample in accordance with said derived combined spectral characteristic values.

14. A method as specified in claim 13 wherein said step of deriving spectral characteristic values comprises deriving at least three sets of spectral characteristic values and deriving said combined spectral characteristic values from said three sets.

42. A method as specified in claim 10, 16, 17, 22, 25, or 32 wherein the steps of said method are repeated after the tissue undergoes medical treatment.

43. A method as specified in claim 10, 16, 17, 22, 25, or 32 wherein the steps of said method are performed on tissue suspected of having disease, wherein the steps are repeated if disease is detected to determine the development of said disease, and wherein the tissue is subjected to medical treatment, and wherein the steps of said method are again repeated to determine the response of said tissue to said medical treatment.

16. A method for processing reflected ultrasonic signals from body tissue and for generating a display of at least a portion of a cross-section of said tissue, representative of clinically significant tissue characteristics, comprising:

transmitting wideband ultrasonic pulse signals into a region of interest of said tissue in an ordered spatial progression corresponding to said cross-section;

receiving reflected ultrasonic signals from said tissue, said signals comprising signal time intervals corresponding to reflections from successive linear acoustic paths through said tissue in said spatial progression;

dividing selected time periods of said received signals into signal time portions, each signal time portion comprising a selected time sample of one of said signal time intervals and corresponding to a spatial sample of said tissue along a segment of one of said acoustic paths in said region of interest;

performing a spectral analysis of each of said signal time portions to derive therefrom spectral data representative of the power level of said signal time portion at various frequencies with the frequency band of said ultrasonic pulse signals;

normalizing said spectral data to values of spectral data for an object with known reflection characteristics;

deriving spectral characteristic values from said normalized spectral data independently for each of said signal time portions corresponding to a segment of an acoustic path in a region of interest of said cross-section, said spectral characteristic value being selected from the group consisting of frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty;

and generating a display of said region of interest of said cross-section, said display having elemental display areas corresponding to said spatial samples, each of said display areas having a display characteristic representative of said spectral characteristic value.

15. A method as specified in claim 16 wherein said step of generating a display comprises generating a display for each spatial region with a color selected in accordance with said derived spectral characteristic value.

17. A method for generating a display at least a portion of a cross-section of eye tissue representative of the presence in said cross-section of at least one known tissue type, comprising:

transmitting ultrasonic signals into said eye tissue in an ordered spatial progression of acoustic paths corresponding to said cross-section;

receiving reflected ultrasonic signals from said tissue comprising signal time intervals corresponding to reflections from said acoustic paths through said tissue;

dividing selected time periods of said received signals into signal time portions, each signal time portion comprising a selected time sample of one of said signal time intervals and corresponding to a spatial sample of said tissue along a segment of one of said acoustic paths in a region of interest of said eye tissue;

performing a spectral analysis of each of said signal time portions to derive therefrom spectral data representative of the power level of said signal time portions at various frequencies within the frequency band of interest of said ultrasonic pulse signals;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

deriving spectral characteristic values from said normalized spectral data independently for each of said signal time portions corresponding to a segment of an acoustic path in a region of interest in said cross-section, said spectral characteristic value being selected from the group consisting of frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty, said selection being made to correspond with distinguishing properties of aaid known tissue types;

and generating a display of said region of interest of said cross-section, said display having elemental display areas corresponding to said spatial sample, each of said display areas having a spectral characteristic value corresponding to said known tissue type being provided with a distinguishing display characteristic;

whereby a predominance of said distinguishing display characteristic in a portion of said display is indicative of the presence of said known tissue type.

18. A method for processing reflected acoustic signals from a body of material and for generating a display of at least a portion of said body, representative of characteristics of said material comprising:

transmitting acoustic signals into a region of interest in said body in an ordered spatial progression;

receiving reflected acoustic signals from said body;

performing a spectral analysis of selected time portions of said received signal to derive spectral data representative of received signal characteristics for spatial samples of said body within said body region of interest, said spatial samples corresponding to said time portions of said received signal;

normalizing said spectral data to values of spectral data for a object with known reflection characteristics;

deriving periodicity data from said normalized spectral data for each of said spatial samples;

deriving periodicity characteristic values from said periodicity data, said periodicity values being selected to correlate with significant characteristics of said material; and

generating a display of said body region of interest with a display characteristic for each spatial sample of said region selected in accordance with said derived periodicity characteristic values.

19. A method as specified in claim 18 wherein said step of deriving periodicity data comprises deriving cepstral data.

20. A method as specified in claim 18 wherein said step of deriving periodicity data comprises deriving spatial correlation function data.

21. A method as specified in claim 18 wherein said step of deriving periodicity characteristic values comprises deriving a value representative of effective scattering particle spacing for each of said samples.

22. A method for processing reflected ultrasonic signals from body tissue material and for generating a display of at least a portion of said tissue, representative of average particle size of said tissue comprising:

transmitting ultrasonic signals into a region of interest of said tissue in an ordered spatial progression;

receiving reflected ultrasonic signals from said tissue;

performing a spectral analysis of selected time portions of said received signal to derive spectral data representative of received signal characteristics for spatial samples of said tissue within said region of interest, said spatial samples corresponding to said time portions of said received signal;

normalizing said spectral data to values of spectral data for an object with known reflection characteristics;

deriving periodicity data from said normalized spectral data for each of said spatial samples;

deriving periodicity characteristic values from said periodicity data, said periodicity values being selected to correlate with average particle size of said tissue; and

generating a display of said body region of interest with a displaycharacteristic for each spatial sample of said region selected in accordance with said derived periodicity characteristic values.

23. A method a specified in claim 22 wherein said step of deriving periodicity data comprises deriving cepstral data.

24. A method as specified in claim 22 wherein said step of deriving periodicity data comprises deriving spatial correlation function data.

25. A method for processing reflected ultrasonic signals from body tissue and for generating a display of said tissue representative of clinically significant tisue characteristic, comprising:

transmitting ultrasonic signals into said tissue in an ordered spatial progression;

receiving reflected ultrasonic signals from said tissue;

performing a spectral analysis of selected time interval portions of said received signal corresponding to spatial samples of said tissue to derive spectral data representative of spectral power as a function of frequency for each spatial sample of said tissue within a tissue region of interest;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

deriving spectral characteristic values from said normalized spectral data independently for each of said spatial samples, said spectral characteristic values being selected from the group comprising frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty;

generating correlation values for each spatial sample by comparing said spectral characteristic values to corresponding values for tissue having known tissue characteristics; and

generating a display of said tissue region representative of said correlation values for each spatial sample.

26. A method as specified in claim 25 wherein said step of deriving spectral characteristic values comprises deriving values representative of spectral amplitude.

27. A method as specified in claim 25 wherein said step of deriving spectral characteristic values comprises deriving values representative of spectral slope.

28. A method as specified in claim 25 wherein said step of deriving spectral characteristic values comprises deriving values representative of spectral amplitude uncertainty.

29. A method as specified in claim 25 wherein said step of deriving spectral characteristic values comprises deriving at least two sets of spectral characteristic values selected from said groups, and deriving combined spectral characteristic values from said at least two sets of spectral characteristic values, said combined spectral characteristic values being derived to correlate with clinically significant tissue characteristics.

30. A method as specified in claim 29 wherein said step of deriving spectral characteristic values comprises deriving at least three sets of characteristic values and deriving said combined spectral characteristic values from said three sets.

31. A method as specified in claim 25 wherein said step of generating a display comprises generating a display for each spatial region with a color selected in accordance with said correlation values.

32. A method for analyzing tissue characteristics comprising:

transmitting ultrasonic signals into said tissue;

receiving reflected ultrasonic signals from said tissue;

dividing said received signals into signal time periods representative of ultrasonic signals reflected from regions of increasing depthin said tissue;

analyzing said received signals for each of said signal time periods to derive spectral data representative of the spectral characteristics for each of said signal time periods;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

analyzing said normalizing spectral data to derive values representative of spectral slope for each of said signal time periods;

analyzing said spectral slope values to define a linear function representative of the variation in spectral slope as a function of depth in said tissue; and

determining the maximum positive deviation of said spectral slope values from said linear function.

33. A method as specified in claim 32 wherein said ultrasonic signals are transmitted to and received from a plurality of adjacent signal paths, and wherein said step of analyzing said received signals includes analyzing signals from a plurality of said paths to derive average spectral values for said paths.

34. In a system for transmitting ultrasonic signals into material and receiving reflected signals for generating a display representative of signal reflecting characteristics of portions of said material, a method for suppressing noise induced images in said display, comprising:

dividing said received signal into time sequential signal samples;

performing a spectral analysis of said signal samples to derive spectral data representative of received spectral characteristics of said samples;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

analyzing said normalized spectral data to derive values of frequency specific normalized spectral amplitudes at various frequencies and spectral slope for each of said samples;

comparing said values of spectral amplitudes and spectral slope to values characteristics of noise and generating a noise representative signal for signal samples having values within a selected range of said noise characteristic values; and

displaying a fixed, selected image in portions of said display corresponding to said signal samples.

35. A method of identifying a tissue type within a region of intrest of body tissue, comprising:

transmitting ultrasonic signals into said tissue;

receiving reflected ultrasonic signals from said region of interest of said tissue;

performing a spectral analysis on said received signal to derive spectral data;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

analyzing said normalized spectral data to derive at least two spectral characteristic values, said values being selected from the group consisting of frequency specific normalized spectral amplitude, spectral slope and spectral amplitude uncertainty;

computing at least one discriminant function from said at least two values, said function being computed according to a formula selected to provide discrimination among tissue types based on spectral characteristic values for signals reflected from tissues having known tissue types;

and identifying said tissue on the basis of the value of said discriminant function.

36. A method as specified in claim 35 wherein said analyzing step incudes deriving at least three spectral characteristics.

37. A method as specified in claim 35 or 36 wherein said computing step includes computing at least two discriminant functions.

38. A method for determining average particle size of scattering particles in a body of material, comprising:

transmitting acoustic signals into said material;

receiving reflected acoustic signals from at least a portion of said material;

analyzing said received signals to derive spectral data representative of spectral characteristics of said reflected signals;

normalizing said spectral data to values of said spectral data for an object with known reflection characteristics;

analyzing said normalized spectral data to derive spectral slope values thereof; and

computing average particle size from said spectral slope values using a formula relating particle size to spectral slope.

39. A method as specified in claim 38 wherein said received signal is divided into signal time samples corresponding to areas of said body of material and wherein said particle size is computed for each of said areas.

40. A method as specified in claim 39 comprising a further step of generating a display having elemental display areas corresponding to said body areas and wherein said display areas are provided with a display characteristic according to said computed average particle size.