Andrew Mackay Breivik, Utah Valley University
Health
The central research question of this project is to determine the reproducibility of high frequency (HF) ultrasonic signals in breast cancer detection. Previous studies on surgical specimens of breast tissue have shown that HF ultrasound (20-80 MHz) appears sensitive to a range of breast pathologies including fibroadenomas, atypical ductal hyperplasia, fibrocystic changes, and carcinomas. A measurement in the ultrasonic signal called the peak density appears most sensitive to the pathology of the breast. The reproducibility of this parameter has not been quantitatively measured in a comprehensive manner. In parallel to a clinical study being conducted at the Huntsman Cancer Institute, we are conducting a laboratory study of the reproducibility of these measurements using chicken and bovine tissue. The ability to reliably determine the pathology of breast tissue with a real-time intra-operative tool would greatly aid in the surgical removal of all malignant tissue, as well as greatly reduce the occurrences of repeat surgeries to remove margins of cancerous tissue that remained. The results of this study will reveal the degree of variability in the signals, thus supplementing previous studies as well as determining the reliability of the results from the current clinical study. The research methodology included the following. Fresh chicken breast and bovine tissue were cut into 4x3x0.5 cm and 4x3x1.5 cm cubes. The tissue was tested at room temperature (23.4oC) using HF ultrasound. Pitch-catch and pulse-echo waveforms were obtained in triplicate measurements of two types: Three measurements with the transducer not leaving the tissue, and three measurements with the transducer lifted off the tissue between measurements. A total of 640 measurements were acquired and analyzed to obtain the spectral peak densities. Preliminary results indicate that the measurements are reproducible to a statistically significant level, thus removing one possible source of uncertainty in the data.