Noise-Robust One-Bit Diffraction Tomography and Optimal Dose Fractionation

This study presents a noise-robust framework for 1-bit diffraction tomography, a novel imaging approach that relies on intensity-only binary measurements obtained through coded apertures. The proposed reconstruction scheme leverages random matrix theory and iterative algorithms, including power iteration and shifted inverse power iteration, to effectively recover 3D object structures under high-noise conditions. Theoretical analysis highlights the de-noising capabilities of the 1-bit scheme, with numerical experiments validating its robustness across varying noise levels, projection densities, and mask configurations. A key contribution is the investigation of dose fractionation, revealing optimal performance at a signal-to-noise ratio near 1, independent of the total dose. This finding addresses the dose-damage trade-off critical in radiation-sensitive imaging applications, such as biological microscopy. The study also explores the spectral properties of the reconstruction process, providing insights into algorithmic convergence and the interplay between eigenvector correlations and spectral gaps.