High-Resolution 3-D Imaging and Tissue Differentiation with Transmission Tomography

Abstract

A three-dimensional High-resolution Ultrasonic Transmission Tomography (HUTT) system has been developed recently under the sponsorship of the Alfred Mann Institute at the University of Southern California that holds the promise of early detection of breast cancer (mm-size lesions) with greater sensitivity (true positives) and specificity (true negatives) than current x-ray mammograghy. In addition to sub-mm resolution in 3-D, the HUTT system has the unique capability of reliable tissue classification by means of the frequency-dependent attenuation characteristics of individual voxels that are extracted from the tomographic data through novel signal processing methods. These methods yield “multi-band signatures” of the various tissue types that are utilized to achieve reliable tissue differentiation via novel segmentation and classification algorithms. The unparalleled high-resolution and tissue differentiation capabilities of the HUTT system have been demonstrated so far with man-made and animal-tissue phantoms. Illustrative results are presented that corroborate these claims, although several challenges remain to make HUTT a clinically acceptable technology. The next critical step is to collect and analyze data from human subjects (female breasts) in order to demonstrate the key capability of the HUTT system to detect breast lesions early (at the mm-size stage) and to differentiate between malignant and benign lesions in a manner that is far superior (in terms of sensitivity and specificity) to the current x-ray mammography. The key initial application of the HUTT imaging technology is envisioned to be the early (at the mm-size) detection of breast cancer, which represents a major threat to the well-being of women around the world. The potential impact is estimated in hundreds of thousands lives saved, millions of unnecessary biopsies avoided, and billions of dollars saved in national health-care costs every year – to say nothing of the tens of thousands of relieved Radiologists worldwide, who will finally have at their disposal a reliable and effective diagnostic tool for early detection of breast cancer. Numerous other potential applications of this medical imaging technology are possible, following proper adjustments to the specific scanning requirements of each particular application