Adversarially learned iterative reconstruction for imaging inverse problems

TL;DR

This paper introduces an unsupervised iterative reconstruction method for imaging inverse problems, particularly in medical imaging, that matches supervised methods in quality without requiring ground-truth data, while avoiding over-smoothing.

Contribution

It presents a novel unsupervised learning framework based on maximum-likelihood principles for inverse problems, improving reconstruction quality over supervised methods.

Findings

Performs comparably to supervised methods in quality metrics

Avoids over-smoothing common in supervised approaches

Requires higher training complexity but similar inference time

Abstract

In numerous practical applications, especially in medical image reconstruction, it is often infeasible to obtain a large ensemble of ground-truth/measurement pairs for supervised learning. Therefore, it is imperative to develop unsupervised learning protocols that are competitive with supervised approaches in performance. Motivated by the maximum-likelihood principle, we propose an unsupervised learning framework for solving ill-posed inverse problems. Instead of seeking pixel-wise proximity between the reconstructed and the ground-truth images, the proposed approach learns an iterative reconstruction network whose output matches the ground-truth in distribution. Considering tomographic reconstruction as an application, we demonstrate that the proposed unsupervised approach not only performs on par with its supervised variant in terms of objective quality measures but also successfully circumvents the issue of over-smoothing that supervised approaches tend to suffer from. The improvement in reconstruction quality comes at the expense of higher training complexity, but, once trained, the reconstruction time remains the same as its supervised counterpart.