Robyn Kira Omer, Utah Valley University
Peak density, which is the number of peaks and valleys in a specified spectral range of high-frequency (HF) ultrasound, correlates to breast pathology in lumpectomy specimens. It has been a question in both previous and current studies, however, whether the thickness of a sample has an independent effect on the peak density. The objective of this study was to discover any correlation, if any, between specimen thickness and peak density in HF ultrasound measurements (10-100 MHz). Phantoms were fabricated from a mixture of water, gelatin, and soluble fiber. Polyethylene microspheres (180-212 micrometer diameter) were embedded into half of the phantom specimens at 0.0003% concentration to simulate tissue heterogeneity. The other phantoms were devoid of microspheres to provide control measurements. Seventy two pitch-catch measurements were acquired in triplicate using 50-MHz transducers, a HF pulser-receiver, and a 1-GHz digital oscilloscope. The waveforms were analyzed to provide spectra and the resulting peak densities were determined. The results indicate that no significant correlation exists between specimen thickness and peak density. The coefficients of correlation for the microsphere and control specimens were 0.366 and 0.652, respectively. The peak density values were most consistent within the control specimens, ranging from 1 to 4. The peak densities for the microsphere phantoms had a greater range of values, varying from 1 to 8. It is believed that the wide variation in peak density for the microsphere phantoms was due to clustering of the microspheres. Future studies will include looking at previous phantom and tissue studies to further investigate the apparent lack of thickness-peak density correlation.