# Feather-Light Fourier Domain Adaptation in Magnetic Resonance Imaging

arXiv: https://arxiv.org/abs/2208.00474 · HTML: https://mlsanity.com/paper/2208.00474/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 28 references that are in the map.
- First posted on arXiv 31 Jul 2022.

## Builds on (most cited first, 18 of 18)

- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [FDA: Fourier Domain Adaptation for Semantic Segmentation](https://mlsanity.com/paper/2004.05498/) (2020)
- [Domain Adaptation for Medical Image Analysis: A Survey](https://mlsanity.com/paper/2102.09508/) (2021)
- [FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space](https://mlsanity.com/paper/2103.06030/) (2021)
- [Unsupervised domain adaptation in brain lesion segmentation with adversarial networks](https://mlsanity.com/paper/1612.08894/) (2016)
- [Unsupervised domain adaptation for medical imaging segmentation with self-ensembling](https://mlsanity.com/paper/1811.06042/) (2018)
- [Exploring the structure of a real-time, arbitrary neural artistic stylization network](https://mlsanity.com/paper/1705.06830/) (2017)
- [Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy](https://mlsanity.com/paper/1809.04430/) (2018)
- [Phase Consistent Ecological Domain Adaptation](https://mlsanity.com/paper/2004.04923/) (2020)
- [FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysis](https://mlsanity.com/paper/2112.01148/) (2021)
- [Generative Adversarial Networks for Image-to-Image Translation on Multi-Contrast MR Images - A Comparison of CycleGAN and UNIT](https://mlsanity.com/paper/1806.07777/) (2018)
- [Neural Style Transfer Improves 3D Cardiovascular MR Image Segmentation on Inconsistent Data](https://mlsanity.com/paper/1909.09716/) (2019)
- [First U-Net Layers Contain More Domain Specific Information Than The Last Ones](https://mlsanity.com/paper/2008.07357/) (2020)
- [AI Fairness via Domain Adaptation](https://mlsanity.com/paper/2104.01109/) (2021)
- [Microscopy Image Restoration with Deep Wiener-Kolmogorov filters](https://mlsanity.com/paper/1911.10989/) (2019)
- [Ultrasound Domain Adaptation Using Frequency Domain Analysis](https://mlsanity.com/paper/2109.09969/) (2021)
- [Global Adaptive Filtering Layer for Computer Vision](https://mlsanity.com/paper/2010.01177/) (2020)

## Built on by (most cited first, 5 of 5)

- [Domain Generalization for Medical Image Analysis: A Review](https://mlsanity.com/paper/2310.08598/) (2023)
- [Fourier Test-time Adaptation with Multi-level Consistency for Robust Classification](https://mlsanity.com/paper/2306.02544/) (2023)
- [A Fourier Transform Framework for Domain Adaptation](https://mlsanity.com/paper/2403.07798/) (2024)
- [M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation](https://mlsanity.com/paper/2502.17029/) (2025)
- [VISTA: Variance-Gated Inter-Sequence Test-Time Adaptation for Multi-Sequence MRI Segmentation](https://mlsanity.com/paper/2605.17433/) (2026)

## Abstract (arXiv)

Generalizability of deep learning models may be severely affected by the difference in the distributions of the train (source domain) and the test (target domain) sets, e.g., when the sets are produced by different hardware. As a consequence of this domain shift, a certain model might perform well on data from one clinic, and then fail when deployed in another. We propose a very light and transparent approach to perform test-time domain adaptation. The idea is to substitute the target low-frequency Fourier space components that are deemed to reflect the style of an image. To maximize the performance, we implement the "optimal style donor" selection technique, and use a number of source data points for altering a single target scan appearance (Multi-Source Transferring). We study the effect of severity of domain shift on the performance of the method, and show that our training-free approach reaches the state-of-the-art level of complicated deep domain adaptation models. The code for our experiments is released.