research Deep material priors Multimodal vision transformer foundation models for electron microscopy, with Toyota Research Institute. Deep generative priors for ptychography Self-supervised neural priors make electron ptychography faster, sharper, and more robust to noise. Quantitative Lorentz TEM from a single image Self-supervised and simulation-trained ML for magnetic phase imaging. Skyrmions in van der Waals magnets Cryogenic and in situ Lorentz TEM of topological spin textures in 2D magnets. Whole-pattern fitting for strain mapping Fitting entire diffraction patterns for more precise 4D-STEM strain maps, with Samsung. software quantEM Open-source Python library for quantitative electron microscopy. PyLorentz Python package for Lorentz TEM simulation, phase reconstruction, and ML-based analysis.
