EECS CSPL SEMINAR SERIES
WINTER TERM 1997


Sriram Krishnan

Sriram Krishnan

Dept. of Electrical Engineering and Computer Science

University of Michigan

sriramk@engin.umich.edu



Thursday, March 13, 1997
4:30-5:30 pm
1003 EECS


Adaptive Aberration Correction in Clinical Ultrasound Imaging

Abstract -
The inability of diagnostic ultrasound to detect low contrast lesions deep inside the body has limited its success in cancer diagnosis. One of the major problems limiting contrast resolution is sound velocity inhomogenieties within the body. For example, in abdominal imaging, differences in index of refraction between tissue and subcutaneous fatty layers can distort propagating wavefronts, resulting in reduced contrast in the image. As a result, low-contrast objects cannot be visualized. In addition, artifacts may be introduced into the image. To correct for these aberrations and restore image contrast, we have developed the Parallel Adaptive Receive Compensation Algorithm (PARCA). PARCA models the effects of aberrations as off-axis scatterers insonified by an imperfect beam. Image artifacts created by these scatterers can then be estimated and compensated. Results from controlled aberration models with tissue mimicking gelatin phantoms as well as clinical abdominal images demonstrate that PARCA can significantly improve images degraded by aberrations.


Biosketch -
Please refer to Mr. Krishnan's homepage found through the link shown above.



return to Previous CSPL Seminars homepage