Authors
Nizar Guezzi, Hyojin Seong, Youngho Seo, Taehoon Her, Jaesok Yu
Published in
Ultrasonics. Volume 171. Pages 108364. Nov 05, 2026. Epub Nov 05, 2026.
Abstract
Ultrasound Localization Microscopy (ULM) enables microvascular imaging beyond the diffraction limit and has recently emerged as a promising tool for high-resolution mapping of microvasculature. However, its practical application remains limited by extremely large data volumes, creating bottlenecks in acquisition, transfer, and storage. In this work, we propose C2-ULM, a dual-component framework that combines compressed receive acquisition with a novel MAP-based beamforming approach, termed Coherent Prior MAP (CP-MAP), to reduce data burden while preserving ULM performance. Receive compression is achieved through constrained random subsampling of the array channels, while the proposed beamformer compensates for the increased sidelobes and noise typically introduced by sparse acquisition. The method exploits cross-angle aperture diversity to construct a coherent prior observation, which mitigates the bias introduced by sparse channel sampling and enables reliable MAP estimation under compressed receive apertures. EXPERIMENTAL: rat brain results demonstrate that the CP-MAP32 configuration of C2-ULM, which activates only 25% of the receive channels-corresponding to 12.5% of the data volume in a conventional four sample per wave (4 spw) acquisition-matches or exceeds the image quality of the fully sampled DAS128 baseline. Compared with DAS128, CP-MAP32 improves contrast, reduces sidelobe levels by approximately 5 dB in Doppler ensembles, preserves all microbubble detections within the region of interest, and maintains ULM spatial resolution with only a minor 2 µm reduction. Velocity estimation remains accurate, with deviations comparable to the full-aperture case. Importantly, the proposed compressed-coherent strategy reduces the data volume to 12.5-25% of conventional acquisition requirements, resulting in substantial improvements in data-transfer efficiency and storage demands. These results demonstrate that the complementary roles of compression and coherence can be jointly exploited to achieve efficient, high-fidelity ULM. From a neuroimaging perspective, the proposed C2-ULM framework provides a practical pathway toward scalable and bandwidth-efficient mapping of microvascular structure and flow, without compromising localization precision, reconstruction quality, or flow quantification.
PMID:
43035060
Bibliographic data and abstract were imported from PubMed on 16 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 4
- Comments 0