Method and apparatus for automatic tracing of Doppler time-velocity waveform envelope

• US 5,935,074 A
• Filed: 10/06/1997
• Issued: 08/10/1999
• Est. Priority Date: 10/06/1997
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: McCann, Kevin, Ishiguro, Shinichi, Mo, Larry Y. L., Honda, Masayoshi, Becker, David D.
• Primary Examiner(s): Lateef, Marvin M.
• Assistant Examiner(s): PATEL, MAULIN M
• Application Number:
  US08/944,119
• Time in Patent Office: 673 Days
• Field of Search: 600/443, 600/447, 600/454, 600/455, 128/916
• US Class Current: 600/454
• CPC Class Codes: G01S 15/8979 Combined Doppler and pulse-echo imaging systems
• CPC Class Codes: G01S 15/8984 Measuring the velocity vector
• CPC Class Codes: G01S 7/52026 Extracting wanted echo signals Doppler systems G01S15/50; Doppler short range imaging systems G01S15/8979

Abstract

An analytical method for automatic tracing of a Doppler time-velocity waveform envelope. At the core of this algorithm is a maximum frequency detection mechanism which is based on searching for the highest frequency bin whose spectral amplitude exceeds a certain noise threshold. The method is based on a theoretical noise amplitude distribution in the video spectral domain. In particular, the method uses a precise model of the statistical distribution of the video spectral power of white noise to establish a threshold for maximum frequency detection. Input to the noise model is the average white noise level in the video spectral display, which can be computed using either of two analytical methods.

Claims

1. A method for automatic tracing of a Doppler velocity-time waveform envelope on a display monitor, comprising the steps of:

transmitting pulses of ultrasound into a sample volume of ultrasound scatterers;

acquiring a multiplicity of successive samples of Doppler signals backscattered from said sample volume of ultrasound scatterers;

processing said Doppler signals to produce spectral line data representing velocity for continuous display of a succession of spectral lines during the continuation of Doppler signal acquisition, the data for each spectral line comprising a multiplicity of frequency bins for different frequency intervals, each bin containing spectral power data for a respective frequency interval, the spectral power data in each bin for each spectral line being displayed as a grey scale in a respective pixel of a corresponding column of pixels on the display monitor;

for each spectral line, estimating a respective mean spectral noise power, determining a respective spectral noise threshold calculated as a function of said respective estimated mean spectral noise level, and estimating a highest frequency for those frequency bins having a spectral power which exceeds said respective spectral noise threshold; and

tracing an envelope through the pixels corresponding to said frequency bins of highest frequency having a spectral power exceeding said respective spectral noise thresholds.

2. The method as defined in claim 1, further comprising the step of filtering the estimated highest frequency over a plurality of successive time points.

3. The method as defined in claim 1, wherein said step of estimating the highest frequency is performed for both positive and negative flow.

4. The method as defined in claim 3, further comprising the step of determining the flow direction having a dominant signal spectrum if a highest frequency is estimated for both positive and negative flow.

5. The method as defined in claim 4, wherein said step of determining the flow direction having a dominant signal spectrum comprises the steps of comparing the integrated sum of spectral amplitudes from zero frequency to said estimated highest frequency for positive flow and the integrated sum of spectral amplitudes from zero frequency to said estimated highest frequency for negative flow, and then using in the tracing the estimated highest frequency for the flow having the largest integrated sum.

6. The method as defined in claim 1, wherein said step of estimating a respective mean spectral noise level comprises the steps of:

computing a total system noise for a given sample volume position;

dividing said total system noise by the number of Fast Fourier Transform points to yield a mean spectral noise power m;

determining an exponential noise distribution f_x (x) as a function of m in accordance with the equation;

##EQU9## transforming f_x (x) into f_z (z) in accordance with the equation;

##EQU10## and determining a mean video spectral noise level in accordance with the equation;

##EQU11## .

7. The method as defined in claim 6, wherein for the case of a linear grey mapping, the mean video spectral noise level is determined in accordance with the equation:

##EQU12## .

8. The method as defined in claim 1, wherein said step of estimating a respective mean spectral noise level comprises the steps of:

averaging the spectral amplitude of the last N_S frequency bins for at least one of the positive and negative flow; and

averaging the averaged spectral amplitude over N_tp time points,

wherein N_S and N_tp are integers.

9. The method as defined in claim 1 wherein said respective spectral noise threshold is determined in accordance with the equation:

##EQU13## .

10. The method as defined in claim 9, wherein curves of the spectral noise threshold versus the mean spectral noise level as a function of dynamic range are pre-calculated and stored in a look-up memory.

11. A system for automatic tracing of a Doppler velocity-time waveform envelope on a display monitor, comprising:

means for transmitting pulses of ultrasound into a sample volume of ultrasound scatterers;

means for acquiring a multiplicity of successive samples of Doppler signals backscattered from said sample volume of ultrasound scatterers;

a display monitor comprising columns of pixels;

means for processing said Doppler signals to produce spectral line data representing velocity for continuous display of a succession of spectral lines during the continuation of Doppler signal acquisition, the data for each spectral line comprising a multiplicity of frequency bins for different frequency intervals, each bin containing spectral power data for a respective frequency interval, the spectral power data in each bin for each spectral line being displayed as a grey scale in a respective pixel of a corresponding column of pixels on said display monitor;

means for estimating a respective mean spectral noise power for each spectral line;

means for determining a respective spectral noise threshold for each spectral line, calculated as a function of said respective estimated mean spectral noise level;

means for estimating for each spectral line a highest frequency for those frequency bins having a spectral power which exceeds said respective spectral noise threshold; and

means for tracing an envelope through the pixels corresponding to said frequency bins of highest frequency having a spectral power exceeding said respective spectral noise thresholds.

12. The system as defined in claim 11, further comprising means for filtering the estimated highest frequency over a plurality of successive time points.

13. The system as defined in claim 11, wherein said means for estimating the highest frequency comprise means for estimating the highest frequency for both positive and negative flow.

14. The system as defined in claim 13, further comprising means for determining the flow direction having a dominant signal spectrum if a highest frequency is estimated for both positive and negative flow.

15. The system as defined in claim 14, wherein said means for determining the flow direction having a dominant signal spectrum comprise means for comparing the integrated sum of spectral amplitudes from zero frequency to said estimated highest frequency for positive flow and the integrated sum of spectral amplitudes from zero frequency to said estimated highest frequency for negative flow, and said means for tracing uses the estimated highest frequency for the flow having the largest integrated sum.

16. The system as defined in claim 11, wherein said means for estimating a respective mean spectral noise level comprise:

means for computing a total system noise for a given sample volume position;

means for dividing said total system noise by the number of Fast Fourier Transform points to yield a mean spectral noise power m;

means for determining an exponential noise distribution f_x (x) as a function of m in accordance with the equation;

##EQU14## means for transforming f_x (x) into f_z (z) in accordance with the equation;

##EQU15## and means for determining a mean video spectral noise level in accordance with the equation;

##EQU16## .

17. The system as defined in claim 16, wherein for the case of a linear grey mapping, the mean video spectral noise level is determined in accordance with the equation:

##EQU17## .

18. The system as defined in claim 11, wherein said means for estimating a respective mean spectral noise level comprise:

means for averaging the spectral amplitude of the last N_S frequency bins for at least one of the positive and negative flow; and

means for averaging the averaged spectral amplitude over N_tp time points,

wherein N_S and N_tp are integers.

19. The system as defined in claim 11, wherein said respective spectral noise threshold is determined in accordance with the equation:

##EQU18## .

20. The system as defined in claim 19, further comprising look-up memory means for storing curves of the spectral noise threshold versus the mean spectral noise level as a function of dynamic range.