Phase Retrieval with Holography and Untrained Priors: Tackling the Challenges of Low-Photon Nanoscale Imaging
This work provides a more robust phase retrieval method for scientists imaging nanoscale specimens like viruses and proteins under challenging low-photon conditions, offering better resilience to noise and data imperfections.
This paper addresses holographic phase retrieval, a method for reconstructing signals from magnitude-only Fourier measurements, particularly in low-photon nanoscale imaging where data is corrupted by Poisson noise and lacks low-frequency content. The authors introduce a dataset-free deep learning framework that incorporates the physical forward model, a Poisson log-likelihood objective, and an optional untrained deep image prior, demonstrating improved signal recovery from higher noise levels and resilience to suboptimal reference design and missing low-frequency data compared to classical methods.
Phase retrieval is the inverse problem of recovering a signal from magnitude-only Fourier measurements, and underlies numerous imaging modalities, such as Coherent Diffraction Imaging (CDI). A variant of this setup, known as holography, includes a reference object that is placed adjacent to the specimen of interest before measurements are collected. The resulting inverse problem, known as holographic phase retrieval, is well-known to have improved problem conditioning relative to the original. This innovation, i.e. Holographic CDI, becomes crucial at the nanoscale, where imaging specimens such as viruses, proteins, and crystals require low-photon measurements. This data is highly corrupted by Poisson shot noise, and often lacks low-frequency content as well. In this work, we introduce a dataset-free deep learning framework for holographic phase retrieval adapted to these challenges. The key ingredients of our approach are the explicit and flexible incorporation of the physical forward model into an automatic differentiation procedure, the Poisson log-likelihood objective function, and an optional untrained deep image prior. We perform extensive evaluation under realistic conditions. Compared to competing classical methods, our method recovers signal from higher noise levels and is more resilient to suboptimal reference design, as well as to large missing regions of low frequencies in the observations. Finally, we show that these properties carry over to experimental data acquired on optical wavelengths. To the best of our knowledge, this is the first work to consider a dataset-free machine learning approach for holographic phase retrieval.