Super-resolution ultrasound imaging in heterogeneous media via multi-focal blind acoustical structured illumination

In medical ultrasound imaging, breaking the diffraction limit to achieve super-resolution imaging has long been a research focus. Acoustical structured illumination (ASI) technology projects a pre-defined multi-focus acoustic field to shift high-frequency information into the system's passband, demonstrating the potential for resolution enhancement. However, the reconstruction quality of ASI heavily relies on the precise prior knowledge of the illumination field. In practical clinical applications, especially in heterogeneous tissues, the acoustic velocity inhomogeneity of human tissues can lead to severe wavefront distortion and phase aberrations, disrupting the intended illumination pattern and causing traditional ASI reconstruction to fail with serious image artifacts. To address these challenges of ASI in heterogeneous media, this study proposes a multi-focus blind acoustical structured illumination (B-ASI) imaging method. This approach no longer depends on the precise prior acoustic field information. Instead, it utilizes the statistical characteristics of the illumination patterns and combines them with an iterative optimization framework to blindly compute super-resolved images under conditions of unknown field distortion. Simulation and phantom experiments show that under heterogeneous medium conditions, mismatch between the decoding acoustic field and the actual acoustic field leads to decoding errors in ASI. In contrast to the failure of traditional ASI methods, B-ASI effectively leverages the structural characteristics still present in the actual acoustic field to circumvent decoding failures caused by aberrations, thereby successfully resolving closely spaced point targets. This study provides a practical and effective solution for high-resolution ultrasonic structured acoustic illumination imaging in complex imaging media environments.