High resolution Image reconstruction for Passive Ultrasound Imaging of Cavitation
Funding: CAVIIAR Project (ANR-22-CE19-0006), operated by the French National Research Agency (ANR).
Background image credit: © Tatiana Gelvez
Description: In this project, we aimed to develop advanced model-based and learning-based methods to improve medical image restoration for Passive Acoustic Mapping. Specifically, we address temporal resolution limitations to enhance real-time monitoring of ultrasound cavitation by increasing resolution through a convolutional model-based method formulated in the time domain, providing more precise and efficient methods for therapeutic ultrasound applications.
Personal contribution: I contributed to the development of the proposed convolutional beamforming framework for Passive Acoustic Mapping (TD-CM-PAM), from the initial methodological conception to the complete validation of the approach. This included formulating the time-domain convolutional model, designing the associated inverse problem and regularized reconstruction algorithm, and conducting validation. The proposed method demonstrated improved temporal resolution and computational efficiency compared to conventional beamforming techniques.
Associated publications:
Efficient Convolutional Forward Model for Passive Acoustic Mapping and Temporal Monitoring.
Submitted to Signal Processing Letters. arXiv preprint arXiv:2601.07356 (2026)
Gelvez-Barrera, T., Nicolas, B., Gilles, B., Basarab, A. Kouamé, D.