压缩感知在合成发射孔径医学超声成像中的应用

为了解决合成发射孔径技术在医学超声成像实现中面临的数据量大及接收通道多的问题,提出一种超声成像系统频率域稀疏性模型的压缩感知成像算法。首先对超声系统频率域稀疏性模型进行了验证;然后根据稀疏性模型利用压缩感知理论对回波信号进行压缩采样,并使用最优化方法完成回波信号重建;最终通过合成发射孔径技术完成超声成像。针对医学成像中常用的点目标及模拟胎儿目标进行成像仿真实验,对重建图像在均方误差、分辨率及成像质量等方面与常规成像结果对比分析。实验结果表明在保证成像质量的同时,仅使用30%原始数据量及50%总接收通道数目可完成成像;频率域稀疏性模型的压缩感知成像算法可以大幅度减少合成发射孔径成像所需数据量及接收通道数,极大地降低了系统复杂度。 

The application of compressed sensing in synthetic transmit aperture medical ultrasound imaging

In synthetic transmit aperture medical ultrasound imaging field,a compressed sensing ultrasound imaging method based on the sparsity in frequency domain is presented in order to reduce the huge amount of data and huge amount of reciving channels.First,the sparsity in frequency domain was verified.Then the echo signal was compressively sampled in time-spatial domain based on compressive sensing and the echo signal was reconstructed by solving an optimization problem.Finally the image was made by using synthetic transmit aperture approach.The experiments based on point target and fetus target were used to verify the proposed method.The RMS error,resolution and image quality of reconstructed image and those of original image were compared and analyzed.The results showed that only 30%amount of data and 50%of receiving channels were used to implement ultrasound imaging without reducing the quality of image in the experiment.The amount of data and the complexity of system are reduced greatly by the proposed method based on compressed sensing method.