Abstract
Ultrasound is an attractive modality for monitoring temperature during heating of tumors to therapeutic levels. Previously, we predicted monotonic changes in ultrasonic backscattered energy (CBE) for certain sub-wavelength scatterers. Accurate measurement of CBE requires compensation for apparent motion of image features. We measured CBE in 2D in motion-compensated images of four 1-cm thick samples of bovine liver, two of turkey breast, and one of pork muscle during heating in a water bath from 37 to 50oC. Images were formed by a Terason 2000 imager with a 7 MHz linear probe focused at 4.5 cm, the center of each tissue specimen. Employing RF signals from the Terason 2000 (courtesy Teratech Corp.) permitted the use of cross-correlation as a similarity measure for automatic feature tracking with temperature. Tissue motion in 8 image regions of each specimen was tracked from 37 to 50oC in 0.5oC steps. Maximum displacement in all specimens was about 0.5 mm in both axial and lateral directions. Motion compensated image regions were demodulated and smoothed. Pixel values were squared to form the backscattered energy. We compared means of both the positive and negative changes in the BE images. CBE was monotonic. BE differed by about 4 dB at 50oC from its value at 37oC. We measured motion in 3D in one sample of turkey breast by taking 7 images at each temperature displaced by 0.6 mm in elevation. We plan to extend our 2D methods to 3D motion compensation, but because beam width in elevation was 2.5 mm, five times the maximum apparent motion in 2D, effects of apparent motion in elevation on CBE may be negligible. Relatively noise-free CBE curves from tissue volumes of less than 1 cm3 supports the use of CBE for temperature estimation.
Acknowledgement: Supported in part by NIH grant R21-CA90531 from the National Cancer Institute and the Wilkinson Trust at Washington University in St. Louis.