Random Matrix Theory-Based Wiener Postfilter Combined With Eigenspace and Spatial Coherence-Based Minimum Variance Beamformer for Coherent Plane-Wave Adaptive Compounding
Yadan Wang,Shuai Feng,Jingwen Pan,Chichao Zheng,Hu Peng,Yuanguo Wang
DOI: https://doi.org/10.1109/tim.2024.3369145
IF: 5.6
2024-03-12
IEEE Transactions on Instrumentation and Measurement
Abstract:Adaptive beamforming technology is an effective approach to improve the image quality of coherent plane-wave compounding (CPWC). However, when the number of tilted plane waves (PWs) decreases, adaptive beamforming methods usually entail a significant tradeoff between contrast improvement and speckle preservation. In addressing this challenge, we propose a random matrix theory-based Wiener (RMTW) postfilter combined with eigenspace and spatial coherence-based minimum variance (ESCMV) beamformer. The RMTW estimates the signal power of the RMTW by summing the output obtained by combinatorially multiplying the angular dimension data and using the random matrix theory to calculate the noise covariance for noise power estimation. Moreover, we propose an adaptive approach to estimate the eigenvalue threshold based on spatial coherence, aiming to reduce the artifacts induced by eigenspace-based minimum variance (EIBMV). The final image is derived by multiplying the outputs of RMTW and ESCMV. Results show that RMTW–ESCMV with 25 PWs achieves higher image quality than EIBMV with 75 PWs. Specifically, RMTW–ESCMV with 25 PWs averagely improves the contrast ratio (CR) by 23.3 dB (76%) and generalized contrast-to-noise ratio (gCNR) by 0.3 (55%) compared with EIBMV with 75 PWs in the experiment, respectively. For in vivo carotid data, RMTW–ESCMV with 25 PWs provides improvements of 38 dB (93%) in CR and 0.33 (91%) in gCNR, compared with EIBMV with 75 PWs. These findings affirm that the RMTW–ESCMV is a promising method for obtaining high-quality images at high frame rates.
engineering, electrical & electronic,instruments & instrumentation