Michael P. Andr, mandre@ucsd.edu
Helmar S. Jane, Linda K. Olson, Constance D. Lehman, Barbara A. Fecht,
Tuan Nguyen
University of California
SDVA Department of Radiology 114
3350 La Jolla Village Drive
San Diego, CA 92161
Popular version of paper 2aBB2
Presented Tuesday morning, November 2, 1999
138th ASA
Meeting, Columbus, Ohio
Breast cancer can be detected with x-ray mammography at a very early stage when it is still too small to be felt palpably by the women or her doctor. This is why mammography is such an important screening tool in the battle to improve survival of breast cancer.
However, the findings on mammography are frequently uncertain and require the acquisition of additional images--called a diagnostic exam--using ultrasound. Ultrasound may also be used to guide a needle that will obtain a biopsy of the suspicious tissue for lab analysis. The ultrasound picture is formed similar to a radar picture by spraying out a rapid sequence of sound pulses at a frequency much higher than the range of human hearing ("ultra" sound). These pulses are reflected back to the instrument by discontinuities in the breast tissue. The longer the pulse takes to arrive back at its starting point, the farther away is the reflector.
Today's medical ultrasound is a highly sophisticated tool but the underlying technology has not fundamentally changed in more than 10 years. Furthermore, the instruments are not specialized in any way for breast imaging--the same unit may be used to detect liver cancer or vascular clots or examine a baby in the womb.
Current breast ultrasound is one of the most difficult medical imaging procedures and it has several technical limitations. For example, in order to have high resolution, the instrument views only a small region of the breast at a time--a process that has been likened to searching for a marble in a dark warehouse with a flashlight. It requires considerable expertise to perform and the results may vary from operator to operator.
For many years our group has been exploring unusual ways to produce ultrasound pictures of the breast. In particular, we have constructed instruments that utilize portions of the ultrasound beam that are not used by conventional systems, namely the sound that is scattered and transmitted throughout the breast, not just straight back to the point of origin. With this approach we have added to our understanding of how sound interacts with breast tissue and how much that tissue may vary among different women and different diseases.
This talk will describe two such systems and show some images produced from selected patients. Both devices provide a very large field of view that encompasses the entire breast but they employ very different methods. One device uses a circular array of sound detectors and sources that surround the entire breast. It produces a computer image of a slice through the breast--an ultrasound CT scan--using a series of complex computations to extract information about properties of the tissue. A sequence of slices may be stacked up to form a three-dimensional image of the entire breast volume.
The second imaging system involves a principle known as acoustical holography in which scattered sound waves are converted to a visible image by an optical laser beam. Unlike the tomographic device in which the computer may take many minutes to produce a picture, the acoustical holograms appear instantly in a video format that appears surprisingly like an x-ray fluoroscope. Since the images are "real time," a physician may place his hands in the beam to search for a lump or manipulate a needle for a biopsy, but without the safety concerns of fluoroscopy.