Estimating the spatial autocorrelation function for ultrasound scatterers in isotropic media

Jiang Feng Chen, James A. Zagzebski*, Fang Dong, Ernest L. Madsen

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

20 Scopus citations


The autocorrelation function pertaining to spatial distributions of ultrasonic scatterers in soft tissue is believed to contain useful information related to tissue morphology. A simple processing method applied to radio-frequency echo signals estimates this function for a sample having isotropic scattering conditions. It utilizes backscattered echo signals from the sample and echo signals from a reference object having defined scattering properties. The ratio of the echo signal power spectrum from the sample to the echo signal power spectrum from the reference object is obtained, and corrected for attenuation differences between the two media. This yields a 'form factor' for the sample, whose inverse Fourier transform is the autocorrelation function. The method was tested using tissue-mimicking samples for which spatial autocorrelation functions could be modeled from the dimensions of embedded scatterers. The shapes of the measured autocorrelation functions were in reasonable agreement with those estimated, although measured functions overestimated the function at small lag distances. Scatterer diameters estimated from the zeros of the autocorrelation function agreed to within 6% of expected values when the measurement system bandwidth satisfied minimal criteria.

Original languageEnglish (US)
Pages (from-to)648-655
Number of pages8
JournalMedical Physics
Issue number5
StatePublished - May 1998


  • Autocorrelation function
  • Form factor
  • Phantoms
  • Scatterer size
  • Scattering
  • Tissue characterization
  • Ultrasound

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Estimating the spatial autocorrelation function for ultrasound scatterers in isotropic media'. Together they form a unique fingerprint.

Cite this