Ultrasound molecular imaging (UMI) is a new imaging modality that uses targeted microbubbles (MBs) to detect diseases and can be used for early cancer detection. The targeted MBs are designed to bind to specific biomarkers and are highly reflective, allowing the visualization of disease at a molecular level using handheld ultrasound. While promising, UMI faces several challenges in clinical translation caused by image degradation associated with the human imaging environment. We have developed new image reconstruction techniques based on non-traditional beamforming and deep learning to substantially increase the sensitivity and specificity of UMI, both for destructive imaging (e.g., differential targeted enhancement), as well as for nondestructive real-time freehand imaging. We are currently investigating the application of UMI for the early detection of breast cancer.

• D. Hyun, L. Abou-Elkacem, V. A. Perez, S. M. Chowdhury, J. K. Willmann and J. J. Dahl, "Improved Sensitivity in Ultrasound Molecular Imaging With Coherence-Based Beamforming," in IEEE Transactions on Medical Imaging, vol. 37, no. 1, pp. 241-250, Jan. 2018, doi: 10.1109/TMI.2017.2774814.
• D. Hyun, L. Abou-Elkacem, R. Bam, L. L. Brickson, C. D. Herickhoff and J. J. Dahl, "Nondestructive Detection of Targeted Microbubbles Using Dual-Mode Data and Deep Learning for Real-Time Ultrasound Molecular Imaging," in IEEE Transactions on Medical Imaging, vol. 39, no. 10, pp. 3079-3088, Oct. 2020, doi: 10.1109/TMI.2020.2986762.