Joint Cryo-ET Alignment and Reconstruction with Neural Deformation Fields

TL;DR

This paper introduces a neural deformation field framework for joint alignment and reconstruction in cryo-electron tomography, improving the recovery of high-frequency details by automatically modeling deformations during imaging.

Contribution

It presents a novel neural method that simultaneously estimates deformations and reconstructs volumes, eliminating manual alignment steps in CryoET workflows.

Findings

Enables high-frequency detail recovery in reconstructions.

Automatically aligns projections without manual feedback.

Outperforms existing methods in handling deformations.

Abstract

We propose a framework to jointly determine the deformation parameters and reconstruct the unknown volume in electron cryotomography (CryoET). CryoET aims to reconstruct three-dimensional biological samples from two-dimensional projections. A major challenge is that we can only acquire projections for a limited range of tilts, and that each projection undergoes an unknown deformation during acquisition. Not accounting for these deformations results in poor reconstruction. The existing CryoET software packages attempt to align the projections, often in a workflow which uses manual feedback. Our proposed method sidesteps this inconvenience by automatically computing a set of undeformed projections while simultaneously reconstructing the unknown volume. We achieve this by learning a continuous representation of the undeformed measurements and deformation parameters. We show that our approach enables the recovery of high-frequency details that are destroyed without accounting for deformations.