Background: Focusing errors, or aberration, produced by heterogeneous tissue structures can significantly reduce the spatial and contrast resolution of ultrasonic imaging systems. Aberration can be particularly severe in breast imaging, which has restricted the role of ultrasound in breast cancer diagnosis. In principle, it should be possible to design adaptive focusing methods to correct for the effects of aberration on image quality. Most adaptive focusing methods developed in the late 1980s and 1990s demonstrated limited effectiveness due partly to the use of oversimplified models for aberration in tissue. Several investigators are now developing inverse filter-based focusing methods that are less reliant on assumptions about the spatial organization of aberrating structures in tissue, but more appropriate aberration models are still needed to evaluate new focusing methods in a manner that will effectively guide improvements in their clinical performance.
Objectives: We intend to bridge this conceptual gap using large-scale simulations of ultrasonic wave propagation and image formation. The initial goal is to advance understanding of the physical sources of aberration in breast imaging. Numerical aberration and tumour models will be developed and used to assess the performance of competing adaptive focusing algorithms in increasing sensitivity to several types of breast tumours. The models are expected to contribute to the development of focusing techniques that will improve the diagnostic utility of scans obtained from technically difficult patients.
Approach: The simulations employ a "k-space" algorithm for computations of ultrasound propagation that was developed by Robert Waag and colleagues at the University of Rochester and a three-dimensional model of breast anatomy (see the screen capture from the model in the Image Gallery) that was adapted from similar models for x-ray mammography simulation. The anatomy model uses spline surfaces and fractal structures to represent the architecture of the lactiferous ducts, mammary fat lobules, skin, and supporting connective tissue. The propagation simulator iterates pressure and particle velocity fields to compute ultrasonic scattering from structures defined by the anatomy model. This software can be used to simulate two-dimensional B-mode imaging with a linear array transducer. We are currently in the process of verifying that the synthesized images reproduce the speckle statistics observed in clinical breast images. Planned software enhancements include extension from two-dimensional to three-dimensional imaging simulations, incorporation of nonlinear acoustic interactions in the propagation simulator, and addition of specific types of tumours to the anatomical model.
Recent Conference Presentations:
Y.-T. Shen and J.C. Lacefield, "First-order speckle statistics of ultrasound breast images synthesized from a computational anatomy model," Canadian Acoustical Association Acoustics Week in Canada 2005, Can. Acoust., vol. 33, no. 3, pp. 86-87, 2005.
Y.-T. Shen and J.C. Lacefield, "Computational synthesis of ultrasound breast images from a three-dimensional anatomical model," 149th Meeting of the Acoustical Society of America, J. Acoust. Soc. Am., vol. 117, p. 2444, 2005.
J.C. Lacefield, "Computational evidence for a discrete-scatterer aberration model in medical ultrasound," 147th Meeting of the Acoustical Society of America, J. Acoust. Soc. Am., vol. 115, p. 2522, 2004.
