Background: Ultrasound is a widely used medical imaging modality due to its safety, real-time capabilities, and affordability. However, conventional single-transducer systems are limited in spatial resolution, field-of-view, and image quality, which can hinder detailed visualization of complex anatomical structures. An emerging approach, multi-transducer ultrasound [1-3], offers the potential to overcome these limitations by combining signals from multiple probes, but it introduces significant challenges in data fusion, synchronization, and image reconstruction [4,5].
Novelty & Importance: This project proposes a novel AI-enabled framework to harness the power of multi-transducer ultrasound, integrating advanced deep learning and physics-informed methods for coherent image reconstruction. By focusing on robustness across diverse imaging scenarios, the research addresses critical barriers in the practical deployment of multi-transducer systems. The outcome will be a step-change in ultrasound imaging capabilities, enabling higher-resolution, wide-field imaging that could benefit a broad range of clinical applications, from organ assessment to procedural guidance.
Aims & Objectives: The overarching aim is to develop and validate an AI-enabled multi-transducer ultrasound imaging system that enhances image resolution, field-of-view, and quality compared to conventional approaches. The research objectives are: (1) to develop algorithms for coherent data fusion and high-resolution image reconstruction from multiple transducers, (2) to ensure model robustness and generalization across variations in probe configurations, tissue types, and acoustic conditions, and (3) to experimentally validate the system using simulated, phantom, and in vivo datasets, quantifying improvements in spatial coherence, resolution, and computational efficiency.
This project combines simulation, AI algorithm development, and experimental evaluation to advance the next generation of ultrasound imaging, offering a transformative approach that could redefine non-invasive diagnostic imaging.
[1] Peralta, L. et al. Coherent Multi-Transducer Ultrasound Imaging. IEEE Trans Ultrason Ferroelectr Freq Control 66, 1316–1330 (2019).
[2] Foiret, J. et al. Improving plane wave ultrasound imaging through real-time beamformation across multiple arrays. Scientific Reports 2022 12:1 12, 1–14 (2022).
[3] Peralta, L. et al. 3-D Coherent Multitransducer Ultrasound Imaging With Sparse Spiral Arrays. IEEE Trans Ultrason Ferroelectr Freq Control 70, 197–206 (2023).
[4] Peralta, L. et al. On the Arrays Distribution, Scan Sequence and Apodization in Coherent Dual-Array Ultrasound Imaging Systems. Applied Sciences 2023, Vol. 13, Page 10924 13, 10924 (2023).
[5] Dryburgh, P. et al. Specific Filtered Delay Multiply and Sum beamforming for coherent multi-transducer ultrasound imaging. Ultrasonics 155, 107731 (2025).