Medical ultrasonic imaging has been utilized in a variety of clinical diagnoses for many years. Recently, because of the\nneeds of portable and mobile medical ultrasonic diagnoses, the development of real-time medical ultrasonic imaging\nalgorithms on embedded computing platforms is a rising research direction. Typically, delay-and-sum beamforming\nalgorithm is implemented on embedded medical ultrasonic scanners. Such algorithm is the easiest to implement at\nreal-time frame rate, but the image quality of this algorithm is not high enough for complicated diagnostic cases. As a\nresult, minimum-variance adaptive beamforming algorithm for medical ultrasonic imaging is considered in this paper,\nwhich shows much higher image quality than that of delay-and-sum beamforming algorithm. However, minimumvariance\nadaptive beamforming algorithm is a complicated algorithm with O(n3) computational complexity.\nConsequently, it is not easy to implement such algorithm on embedded computing platform at real-time frame rate.\nOn the other hand, GPU is a well-known parallel computing platform for image processing. Therefore, embedded\nGPU computing platform is considered as a potential real-time implementation platform of minimum-variance\nbeamforming algorithm in this paper. By applying the described effective implementation strategies, the GPU\nimplementation of minimum-variance beamforming algorithm performed more than 100 times faster than the ARM\nimplementation on the same heterogeneous embedded platform. Furthermore, platform power consumptions,\ncomputation energy efficiency, and platform cost efficiency of the experimental heterogeneous embedded platforms\nwere also evaluated, which demonstrated that the investigated heterogeneous embedded computing platforms\nwere suitable for real-time portable or mobile high-quality medical ultrasonic imaging device constructions.
Loading....