Harmonizing Flows: Unsupervised MR harmonization based on normalizing flows

TL;DR

This paper introduces an unsupervised, source-free framework using normalizing flows to harmonize MRI images across domains, improving cross-site brain MRI segmentation performance.

Contribution

It presents a novel three-step unsupervised framework combining a shallow harmonizer and normalizing flows for MRI harmonization without source data.

Findings

Outperforms existing harmonization methods in MRI segmentation tasks

Effective in cross-domain brain MRI segmentation across four sites

Unsupervised and task-independent approach

Abstract

In this paper, we propose an unsupervised framework based on normalizing flows that harmonizes MR images to mimic the distribution of the source domain. The proposed framework consists of three steps. First, a shallow harmonizer network is trained to recover images of the source domain from their augmented versions. A normalizing flow network is then trained to learn the distribution of the source domain. Finally, at test time, a harmonizer network is modified so that the output images match the source domain's distribution learned by the normalizing flow model. Our unsupervised, source-free and task-independent approach is evaluated on cross-domain brain MRI segmentation using data from four different sites. Results demonstrate its superior performance compared to existing methods.