Optimized GPU Framework for Pulsed Wave Doppler Ultrasound

摘要：

Pulsed Wave (PW) spectrum Doppler ultrasound is avaluable tool for clinical diagnosis for flow velocity distributionin vessels. However, real-time processing of PW spectrum iscomputationally intensive, involving wall filtering, Fast FourierTransform (FFT), column filtering and linear averaging. In thispaper a very efficient implementation of a PW Doppler spectrumultrasound using the Compute Unified Device Architecture(CUDA?) platform developed by NVIDIA? Is presented. Byexploiting the explicit parallelism exposed in the graphicshardware we obtain more than one order speed-up gaincompared with that from standard CPUs. Finally, we get a rateof 7.60 μ s with one line of 256 samples, which is about 92 timesfaster than the CPU implementation.

作者: Li Wang Dan Shi Dong C.Liu

作者单位: Computer Science College Sichuan University Chengdu, China

会议类型: 国际会议

会议名称: The 4th International Conference on Bioinformatics and Biomedical Engineering(第四届IEEE生物信息与生物医学工程国际会议 iCBBE 2010)

会议地点: 成都

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-18（万方平台首次上网日期，不代表论文的发表时间）