Implementing Delay Multiply and Sum Beamformer on a Hybrid CPU-GPU Platform for Medical Ultrasound Imaging Using Open MP and CUDA

Anovel beamforming algorithmnamed Delay Multiply and Sum(DMAS),which excels at enhancing the resolution and contrast of ultrasonic image,has recently been proposed.However,there are nested loops in this algorithm,so the calculation complexity is higher compared to the Delay and Sum(DAS)beamformer which is widely used in industry.Thus,we proposed a simple vector-based method to lower its complexity.The key point is to transform the nested loops into several vector operations,which can be efficiently implemented on many parallel platforms,such as Graphics Processing Units(GPUs),and multi-core Central Processing Units(CPUs).Consequently,we considered to implement this algorithm on such a platform.In order to maximize the use of computing power,we use the GPUs andmulti-core CPUs inmixture.The platform used in our test is a low cost Personal Computer(PC),where a GPU and a multi-core CPU are installed.The results show that the hybrid use of a CPU and a GPU can get a significant performance improvement in comparison with using a GPU or using amulti-core CPU alone.The performance of the hybrid system is increased by about 47%–63%compared to a single GPU.When 32 elements are used in receiving,the fame rate basically can reach 30 fps.In the best case,the frame rate can be increased to 40 fps.

Robust Needle Localization and Enhancement Algorithm for Ultrasound by Deep Learning and Beam Steering Methods

Ultrasound(US)imaging is clinically used to guide needle insertions because it is safe,real-time,and low cost.The localization of the needle in the ultrasound image,however,remains a challenging problem due to specular reflection off the smooth surface of the needle,speckle noise,and similar line-like anatomical features.This study presents a novel robust needle localization and enhancement algorithm based on deep learning and beam steering methods with three key innovations.First,we employ beam steering to maximize the reflection intensity of the needle,which can help us to detect and locate the needle precisely.Second,we modify the U-Net which is an end-to-end network commonly used in biomedical segmentation by using two branches instead of one in the last up-sampling layer and adding three layers after the last down-sample layer.Thus,the modified U-Net can real-time segment the needle shaft region,detect the needle tip landmark location and determine whether an image frame contains the needle by one shot.Third,we develop a needle fusion framework that employs the outputs of the multi-task deep learning(MTL)framework to precisely locate the needle tip and enhance needle shaft visualization.Thus,the proposed algorithm can not only greatly reduce the processing time,but also significantly increase the needle localization accuracy and enhance the needle visualization for real-time clinical intervention applications.