Clearer Ultrasound Images Thanks to Mathematics

August 1, 2007

A biomedical engineering team from the University of Virginia has come up with an imaging algorithm that they say transforms for the better the way medical science sees inside our bodies.

Screening for life-threatening conditions using ultrasound technology necessitates images that are well-defined. Yet clinical imaging specialists are aware that, in many cases, patients "image poorly." In other words, images of their organs and tissues aren't clear.

The reason is that off-axis signals — reflections coming from undesired locations — often degrade images produced by current ultrasound systems. The new algorithm allows the team's imaging system to reduce the contribution of these unwanted signals. The resulting images have greatly increased contrast and resolution.

The new algorithm, developed by William F. Walker, Francesco Viola, and Michael A. Ellis, has been dubbed the Time-domain Optimized Near-field Estimator (TONE).

The team tested their algorithm in a series of simulations using sample ultrasound data, then compared their data to conventional beamforming strategies (CBF) used by current ultrasound scanners.

Imaging trials were conducted using wires suspended in water, a typical setup to test image resolution and contrast in medical ultrasound. The results showed improvement in spatial resolution over CBF.

"The potential applications for this algorithm are almost infinite," said James H. Aylor, dean of Virginia's School of Engineering and Applied Science. "Not only can it be used in the medical community to benefit patients nationwide, but it will also have applications in the fields of radio astronomy, seismology, and more."

Source: University of Virginia, July 11, 2007