Method and system for automatic measurements of doppler waveforms

• US 5,647,366 A
• Filed: 09/17/1996
• Issued: 07/15/1997
• Est. Priority Date: 09/17/1996
• Status: Expired due to Term

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Systems, Inc. (Siemens AG)
• Inventors: Weng, Lee
• Primary Examiner(s): Jaworski, Francis
• Assistant Examiner(s): FIELDS, DERRICK LAMAR
• Application Number:
  US08/718,060
• Time in Patent Office: 301 Days
• Field of Search: 128/661.08, 128/661.09
• US Class Current: 600/453
• CPC Class Codes: G01S 15/58 Velocity or trajectory determination systems; Sense-of-movement determination systems velocity measurement in imaging systems G01S15/8979

Abstract

A method and system of processing Doppler data includes determining a noise component independently of automatic signal analysis. A succession of frequency spectra is formed in response to receiving reflected energy from an object of interest. A Doppler waveform is then derived from the succession of frequency spectra. The execution of noise estimation is limited to time periods between peaks of the Doppler waveform, providing a first level of isolation from the true-signal processing. Implementation of the noise estimation is based upon portions of the frequency spectra that are independent of the portions from which the Doppler waveform is derived. This provides a second level of isolation. The determined noise component is used as a threshold level in subsequent derivation of the Doppler waveform. Preferably, there are multiple noise thresholds that are adaptively utilized to permit application of a short search block during Doppler frequency detection.

Claims

1. A method of processing Doppler data comprising steps of:

forming a plurality of first Doppler signals in response to receiving energy from an object of interest, said received energy having a range of frequency components, each first Doppler signal being indicative of intensities of frequency components;

forming a second Doppler signal from said first Doppler signals such that said second Doppler signal is indicative of values of a time-varying parameter of interest, said second Doppler signal having a succession of first peaks; and

determining a noise component within said second Doppler signal, including limiting implementation of said determination to time periods corresponding to regions between said first peaks of said second Doppler signal and including basing said determination upon first portions of said first Doppler signals independent of second portions upon which said values of said time-varying parameter are based.

2. The method of claim 1 wherein said step of forming first Doppler signals includes forming a succession of frequency spectra and wherein said step of forming said second Doppler signal includes forming a maximum Doppler frequency waveform representative of said succession of frequency spectra.

3. The method of claim 2 further comprising utilizing said determined noise component in said step of forming said maximum Doppler frequency waveform, including applying said determined noise component as a threshold level.

4. The method of claim 3 wherein said step of forming frequency spectra includes defining each frequency spectrum in accordance with intensities of successive frequency bins, said step of applying said determined noise component as a threshold level including determining whether successive frequency bins in a first group of bins have intensities greater than said threshold level.

5. The method of claim 4 wherein applying said determined noise component further includes determining whether successive frequency bins of a second group of bins adjacent to said first group of bins have a maximum intensity exceeding a function of said threshold level plus a selected intensity value within a range of intensities representative of said first group of bins.

6. The method of claim 5 wherein applying said determined noise component further includes temporarily increasing said threshold level following each first peak of said maximum Doppler frequency waveform.

7. The method of claim 1 further comprising a step of automatically detecting said first peaks of said second Doppler signal, including utilizing a conditional median filter having a decaying filter level between successive first peaks.

8. The method of claim 1 wherein said first peaks of said second Doppler signal are positive peaks and wherein said second Doppler signal further includes negative peaks on a side of a baseline opposite to said positive peaks, said method further comprising a step of applying a linear least-mean-squared curve-fitting algorithm to said second Doppler signal.

9. The method of claim 1 further comprising a step of directing ultrasonic energy into a body, said step of forming said first Doppler signals being in response to receiving reflected energy from blood.

10. The method of claim 9 further comprising forming a mean frequency waveform from said first Doppler signals, said second Doppler signal being a maximum frequency waveform, said method further comprising detecting beginnings of cardiac cycles, including applying a least-mean-squared curve-fitting algorithm to a rising edge of said maximum frequency waveform and detecting a crosspoint of said mean frequency waveform with extension of a substantially straight line defined by applying said algorithm.

11. A method of processing Doppler data for medical applications comprising steps of:

forming a series of frequency spectra in response to receiving ultrasonic reflections from an object of interest, each frequency spectrum being a sequence of bins;

determining a first noise threshold indicative of a background noise component of said frequency spectra; and

forming a maximum Doppler frequency waveform based upon said frequency spectra, including searching each frequency spectrum for a maximum frequency utilizing a search block of first and second groups of a small number of bins and including utilizing said noise threshold for each application of said search block, wherein utilizing said noise threshold includes determining whether successive bins of said first group each have an intensity value exceeding said noise threshold and, if so, whether a maximum intensity value of successive bins of said second group exceeds a sum of a function of said noise threshold and a calculated intensity value within a range of said intensity values of said first group.

12. The method of claim 11 wherein said step of forming said maximum Doppler frequency waveform includes selecting a search block to include only four bins.

13. The method of claim 11 wherein said steps of determining said noise threshold and forming a maximum frequency waveform are executed using first and second portions of a plurality of said frequency waveform, respectively, said first and second portions being spaced apart by third portions such that said steps are executed substantially in isolation.

14. The method of claim 11 further comprising a step of automatically detecting peaks of said maximum Doppler frequency waveform, including utilizing a conditional median filter having a decaying filter level within regions between successive peaks.

15. The method of claim 11 further comprising a step of applying a linear least-mean-squared curve-fitting approach to said maximum Doppler frequency waveform to smooth transitions across a baseline.

16. A system for providing automatic Doppler data processing for medical applications comprising:

an ultrasound transducer;

spectra means for generating a series of frequency spectra in response to receiving ultrasound energy at said transducer, said spectra means having a first output of said frequency spectra, each frequency spectrum being indicative of intensities of frequency components of said received ultrasound energy;

waveform means connected to said first output for generating a Doppler waveform of values of a time-varying parameter of interest in response to said frequency spectra, said waveform means having a second output in which said Doppler waveform has a succession of first peaks; and

noise estimator means connected to said second output for determining a noise level during periods between successive first peaks based upon first portions of said frequency spectra remote from second portions upon which said values of said Doppler waveform are based.

17. The system of claim 16 wherein said output of said waveform means is a maximum frequency waveform output representative of maximum frequency as a function of time.

18. The system of claim 16 further comprising a conditional median filter means connected to said second output for applying a decaying filter level to said Doppler waveform between successive first peaks.

19. The system of claim 16 wherein said noise estimator means has a third output connected to said waveform means for applying said determined noise level for selecting said values of said time-varying parameter.