About
I develop machine learning and computer vision methods for inverse problems in imaging, with applications including MRI, CT, x-ray, and electron microscopy.
Much of my work has focused on two questions:
- Which data representations best model different imaging systems and subjects?
- How can we image things that change while we are imaging them?
Currently, I am a postdoc at UC Berkeley with Laura Waller. Previously, I did my PhD at MIT with Polina Golland.
My CV as of 8/2026 is available here.
Selected Projects
For full publication list, see Google Scholar.
A Gaussian Parameterization for Direct Atomic Structure Identification in Electron Tomography
Reformulating atomic electron tomography to optimize Gaussian atoms instead of voxelgrids, improving reconstructions and directly identifying atomic structures.
International Conference on Computational Photography, 2025.
Paper Code Video
Data Consistent Deep Rigid MRI Motion Correction
Using test-time model-based optimization on neural network outputs to produce fast, high quality, physics-consistent motion-corrected MRI reconstructions.
Medical Imaging with Deep Learning, 2023 (Oral; Best Paper).
Paper Code Video MIT News
Joint Frequency and Image Space Learning for MRI Reconstruction and Analysis
A general purpose neural network layer that combines frequency and image space features for correcting artifacts in Fourier imaging.
Machine Learning for Biomedical Imaging, 2022.
Paper Code Video