SIMBA: Scalable Inversion in Optical Tomography using Deep Denoising Priors

TL;DR

SIMBA is a scalable iterative algorithm that combines physics-based modeling and deep neural network priors to enable fast, high-quality 3D optical tomography imaging with reduced computational resources.

Contribution

The paper introduces SIMBA, a novel mini-batch iterative method that scales to large datasets and integrates deep denoising priors with the forward model for improved imaging.

Findings

SIMBA reduces computational load significantly.

High-quality 3D images achieved with less data.

Theoretical convergence established for nonexpansive denoisers.

Abstract

Two features desired in a three-dimensional (3D) optical tomographic image reconstruction algorithm are the ability to reduce imaging artifacts and to do fast processing of large data volumes. Traditional iterative inversion algorithms are impractical in this context due to their heavy computational and memory requirements. We propose and experimentally validate a novel scalable iterative mini-batch algorithm (SIMBA) for fast and high-quality optical tomographic imaging. SIMBA enables high-quality imaging by combining two complementary information sources: the physics of the imaging system characterized by its forward model and the imaging prior characterized by a denoising deep neural net. SIMBA easily scales to very large 3D tomographic datasets by processing only a small subset of measurements at each iteration. We establish the theoretical fixed-point convergence of SIMBA under nonexpansive denoisers for convex data-fidelity terms. We validate SIMBA on both simulated and experimentally collected intensity diffraction tomography (IDT) datasets. Our results show that SIMBA can significantly reduce the computational burden of 3D image formation without sacrificing the imaging quality.