Plane Wave Utrsound Imaging On Deep Learning

Plane wave imaging is a fast ultrasonic imaging method in medical imaging technology.It adopts single plane wave emission,without focusing,and allows high frame rate.However,compared with focused wave ultrasound imaging,the quality of single plane wave imaging will be seriously affected.Traditional adaptive beamformer estimates apodization weight from echo trajectories acquired by multiple sensors to improve imaging quality.In this paper,the plane wave imaging method based on neural network is used to obtain better image quality.The proposed method has been applied to the simulation,real body membrane and carotid artery experiments on the open dataset PICMUS.The contributions and achievements of this paper are as follows.1.Several beamformers such as traditional incoherence delay and sum(IDA),delay and sum(DAS),minimum variance unconstrained distortion(MVDR),coherence factor(CF)and phase coherence factor(PCF)are compared and applied to PICMUS data set to obtain different plane wave imaging results.2.In the experiment,the composite plane wave ultrasound imaging of 1 angle,37 angles and 75 angles were carried out for the above different imaging methods,and the results of different angles were compared.3.To solve the problem of less imaging data sets,we use Filed II simulate train datasets and U-net network structure.The whole beamforming is embedded into convolution network and trained with single point target dataset.The single plane wave scanning data is used as the input,and the focused scanning data is used as the output.4.In order to evaluate the performance of the traditional method and the proposed method,the lateral resolution,contrast(CR)and contrast to noise ratio(CNR)are used to quantify the performance.Experimental results show that this method has better lateral resolution and contrast.The proposed methods are lower than the proposed ones in terms of incoherence(IDA),delay superposition(DAS),minimum variance unconstrained distortion(MVDR),coherence factor(CF),phase coherence factor(PCF).These results verify that the proposed method can not only maintain high frame rate,but also ensure high spatial resolution and contrast.