Joseph Roring, Utah Valley University
Physics
Removal of all cancerous tissue in breast conservation surgery (BCS) is critical to prevent local recurrence. Unfortunately, 30-50% of patients require additional surgery due to failure to resect all the necessary tissue. A real-time method for detecting infected tissue is therefore desirable. Previous studies have shown that the complexity of high-frequency (50 MHz) ultrasonic spectra can be correlated to a range of breast pathologies in BCS. However, the mechanism behind this correlation is still not very well understood. The purpose of this research is to explore the connection between tissue micro-heterogeneity and ultrasonic spectral complexity using breast tissue phantoms, i.e. materials that mimic breast tissue properties and microstructure. A physical basis can then be determined that links ultrasonic measurements to breast tissue pathology. Phantoms were made from a Knox® gelatin base and soluble fiber (Metamucil®). Heterogeneities simulating lobular and ductal components of mammary glands were created through the addition of polyethylene microspheres and nylon fibers. Pitch-catch and pulse-echo waveforms were acquired from the samples using high-frequency ultrasound. The data were analyzed by measuring the number of peaks (the peak density) in the first-order spectrum (Fourier transform of the time-domain waveform) and the slope of the second-order spectrum (two consecutive Fourier transforms of the time-domain waveform). The phantom specimens displayed first-order peak densities that were significantly greater and second-order spectral slopes that were significantly lower than homogeneous control samples. Phantoms with large fibers (250 micrometer diameter) showed the highest peak densities with values greater than 3x those of the controls. The peak density trend of the phantom samples with increased microscopic heterogeneity was consistent with data of breast tissue specimens. These results provide a physical mechanism for the use of these parameters in the imaging of breast tissues with atypical and malignant pathologies.