This Academic-Industrial partnership grant program is focused on the development of a new ultrasound tomography system (designated as the Partnership scanner) for breast cancer detection. Steven A. Johnson at TechniScan, Inc. will be the coordinating PI. This scanner is designed to extend and improve on features of TechniScan's previous breast ultrasound tomographic scanning system (USCT scanner). The scanner provides Inverse Scattering Tomography (IST) quantitative imaging of sound speed and acoustic attenuation. IST images are currently made from 180 forward directed scattering measurements on a receiver that moves in 2 degree steps around 360 degrees. It also moves in 1 mm vertical steps to scan three dimensions. The IST spatial resolution is about 1.5 mm horizontally and 2 mm vertically. New algorithms will be developed and upgraded hardware will be constructed during the first two and half years of the 5 year grant period. The new Partnership scanner will increase the number of inverse scattering receiver transducers from 1536 to 3072 and the number of receiver rows from 8 to 16. The number of IST transmitter rows will be increased from 8 to16 to allow for vertical beam steering. This program will improve existing IST and Reflection Tomography (RT) algorithms over its 5 year funding period. The speed for improved calculation will be provided by four or more NVIDIA graphic processor units (GPU). Each GPU is faster than 16 Intel Pentium cores. The new scanner also provides Reflection Tomography (RT) imaging of refraction corrected (focused) and attenuation calibrated imaging of the collected reflection data. RT data are collected with two horizontal reflection transducers that move in 6 degree steps around 360 degrees. These two RT transducers have focal length of 45 mm and 75 mm to provide a large depth of focus when combined. The RT spatial resolution is about 0.5 mm horizontally and 1.5 mm vertically. These two are complemented by a vertical, compounded, electronic beam-steered transducer, which give 0.5 mm vertical resolution and allow imaging of the chest wall. Further image improvement is provided by soft polymer breast cups that stabilize the breast and control/minimize refraction artifacts. Human studies will commence in the third year at the University of California at San Diego and are intended to expand research on the acoustic properties of breast tissue as a means to classify masses as benign or malignant. Michael Andre will head UCSD work as a PI. The subtlety and complexity of breast tissues and masses in patients are greater than can be achieved in tissue-mimicking phantoms so human studies are essential for evaluating performance of the Partnership Breast Scanner.
This project has high societal relevance because of its positive impact on breast cancer diagnosis and survivability. Present methods of breast cancer detection, such as mammography or classical ultrasound, give a large percent of false positives. Eighty percent of patients sent to biopsy return a benign result. The advanced technology ultrasound tomography scanner has potential to provide more accurate diagnosis, find more cancer and reduce unnecessary biopsies.