Unsupervised Knowledge-Transfer for Learned Image Reconstruction

TL;DR

This paper introduces an unsupervised transfer learning method for image reconstruction that effectively adapts to realistic data without supervision, providing uncertainty estimates and outperforming some supervised methods in medical imaging tasks.

Contribution

A novel unsupervised knowledge-transfer framework for learned image reconstruction within a Bayesian setting, capable of adapting to real data without labeled training pairs.

Findings

Competitive with state-of-the-art methods in CT reconstruction

Improves quality on different test data distributions

Provides uncertainty quantification in reconstructions

Abstract

Deep learning-based image reconstruction approaches have demonstrated impressive empirical performance in many imaging modalities. These approaches usually require a large amount of high-quality paired training data, which is often not available in medical imaging. To circumvent this issue we develop a novel unsupervised knowledge-transfer paradigm for learned reconstruction within a Bayesian framework. The proposed approach learns a reconstruction network in two phases. The first phase trains a reconstruction network with a set of ordered pairs comprising of ground truth images of ellipses and the corresponding simulated measurement data. The second phase fine-tunes the pretrained network to more realistic measurement data without supervision. By construction, the framework is capable of delivering predictive uncertainty information over the reconstructed image. We present extensive experimental results on low-dose and sparse-view computed tomography showing that the approach is competitive with several state-of-the-art supervised and unsupervised reconstruction techniques. Moreover, for test data distributed differently from the training data, the proposed framework can significantly improve reconstruction quality not only visually, but also quantitatively in terms of PSNR and SSIM, when compared with learned methods trained on the synthetic dataset only.