Design Support


Ultrasound (US) elastography measures the deformation of tissue in response to applied force to derive and display its stiffness. It has been applied to various organs and tissues, e.g., as an adjunct tool to conventional US and X-ray mammography in the detection and characterization of breast masses. Real-time US elastography processing is demanding due to several computationally-expensive modules, such as crosscorrelation. Multicore digital signal processors (DSPs) provide increased computational performance with less power consumption and lower cost than other programmable approaches, e.g., with general-purpose CPUs and GPUs. However, optimal algorithm implementations on multicore processors require careful algorithm/architecture analysis and prudent mapping to achieve good performance. In this paper, we present our approach and techniques used in mapping our US elastography algorithm on a multicore DSP, i.e., Texas Instruments TMS320C6472. Our results show that a single C6472 multicore DSP can support real-time US elastography processing.