Magnetic resonance elastography (MRE) is a quantitative imaging technique for assessing soft tissue stiffness. MRE protocols acquire wavefields at multiple frequencies, which are then inverted to estimate the shear wave speed as a proxy for tissue stiffness. Existing neural network-based approaches perform single-frequency reconstructions followed by multifrequency aggregation.
In our paper, "Multifrequency Neural Network-based Wave Inversion in MR Elastography”, we introduce the first multifrequency neural network-based wave inversion in MRE. Trained using synthetically generated data, our approach is generalizable to in vivo data and flexible with respect to the acquisition protocol, enabling its application in a wide variety of settings.
Presenting my paper at MICCAI 2025 is an honor and an incredible opportunity for me. It is a chance to share my research with the scientific community, receive feedback, and exchange with experts in the field.
I am very interested in learning more about advancements in deep learning-based reconstructions in medical imaging, as well as innovative approaches integrating physics into machine learning models. Trustworthy AI is also a field I am looking forward to exploring more at the conference.
At the conference, I am excited to hear presentations on state-of-the-art research in AI for medical imaging. I am also looking forward to networking and sharing ideas with researchers from all over the world. Experiencing South Korea and its culture will additionally be an enriching part of the conference!