Detection of blood perfusion

Paul G. Krause, Ahmed H. Tewfik, James F. Greenleaf

Research output: Contribution to conferencePaperpeer-review


Ultrasound detection of capillary blood flow is difficult as capillary blood flow provides a weak backscattered signal with a small Doppler shift. To aid in the detection of such signals, a new algorithm is introduced exploiting two powerful ideas. First, the receiver is developed by creating subspaces corresponding to expected flow and clutter signals. It then exploits the fact that some of the basis vectors in the flow and clutter subspaces are very nearly perpendicular. By projecting the input data onto the perpendicular components of the flow subspace a very sensitive detector can be realized. Secondly, the detector coherently sums inputs from several angles. In vitro ultrasound data is collected and applied to the detector. The algorithm utilizing multiple angles is shown to be sensitive detecting flow in the experimental data down to 5 mm/sec in a 6 mm diameter channel.

Original languageEnglish (US)
Number of pages5
StatePublished - Dec 1 1999
EventInternational Conference on Image Processing (ICIP'99) - Kobe, Jpn
Duration: Oct 24 1999Oct 28 1999


OtherInternational Conference on Image Processing (ICIP'99)
CityKobe, Jpn

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering


Dive into the research topics of 'Detection of blood perfusion'. Together they form a unique fingerprint.

Cite this