Reducing Black Box Region in ESBMV: A Comparative Study of Frequency and Spatial Compounding

Black Box Regions (BBRs), which represent a significant degradation of Eigenspace Based Minimum Variance (ESBMV) ultrasound imaging, occur mainly due to the result of phase decoherence and acoustic attenuation of heterogeneous media. These artefacts obscure clinically relevant anatomical features and thus destroy diagnostic precision. Although adaptive beamforming has advanced significantly, focused approaches for systematic BBR mitigation are still poorly examined. Accordingly, the current investigation assesses the restorative efficiency of non- multiplicative compounding techniques, namely Frequency Compounding (FC) and Spatial Compounding (SC), in reducing BBR artefacts. Performance assessment was done using Field II simulations and experimental phantom data measured by Speckle Signal minus Noise Ratio (SSNR) and Contrast to Noise Ratio (CNR). Findings prove that SC significantly outperforms FC in restoring the signal integrity of BBRs. By introducing angular diversity, SC effectively decorrelates speckle, especially at moderate steering angles. Conversely, while FC allows somewhat enhanced contrast improvement, the ability of the method to recover signal in areas of low coherence is limited. Collectively, these results show that incorporating SC into the ESBMV framework provides a powerful tool to reduce BBR artefacts and dramatically improve image fidelity for complex clinical applications.