Constrained Adaptive Beamforming for Improved Contrast in Breast Ultrasound

Abstract

Ultrasonic imaging plays an important role as an adjunct to mammography, with an emerging role in breast cancer screening. Ultrasound s real-time nature, lack of ionizing radiation, and relative comfort for the patient make it an attractive imaging choice. Unfortunately, ultrasound image quality is often limited. We hypothesize that bright scatterers seriously degrade ultrasound images by introducing image clutter. In the breast bright off-axis echoes may originate from Cooper s ligaments, structured glandular tissue, calcification, fat-soft tissue interfaces, or other structures. We initially proposed using a variant of the Frost Adaptive Beamformer to reduce clutter, but discovered that this technique is nonoptimal for our application. Extensive literature review led us to modify the Spatial Processing Optimized and Constrained algorithm (SPOC) to yield the Time-Domain Optimized Near-field Estimator (TONE). In simulations and phantom experiments this technique yields dramatic improvements in resolution and contrast. We have also developed alternate methods of image formation (LCLS and QCLS) that are computationally simpler and, while less impressive than TONE, also yield dramatic image improvements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2008
Accession Number
ADA548418

Entities

People

  • William F. Walker

Organizations

  • University of Virginia

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Algorithms
  • Computational Complexity
  • Computational Science
  • Data Acquisition
  • Detectors
  • Diagnostic Imaging
  • Estimators
  • Ionizing Radiation
  • Near Field
  • Photoacoustic Tomography
  • Signal Processing
  • Simulations
  • Time Domain
  • Tomography
  • Two Dimensional
  • Ultrasounds
  • Waves

Fields of Study

  • Medicine
  • Physics

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Medical Imaging.
  • Radar Systems Engineering.