Abstract
Magnetic field strength is a major source of domain shift
in magnetic resonance imaging (MRI), affecting signal-to-noise ratio,
tissue contrast, spatial detail, and the visibility of anatomical bound-
aries. The MRIxFields 2026 challenge investigates this problem through
cross-field MRI translation across acquisitions at 0.1T, 1.5T, 3T, 5T,
and 7T. Its three tasks, Any-to-7T, 0.1T-to-High, and Any-to-Any syn-
thesis, require the generation of target-field image characteristics while
preserving subject-specific anatomy. This problem is particularly chal-
lenging because paired acquisitions of the same subject across multiple
field strengths are rarely available for training. We propose a 3D un-
paired cross-field MRI translation framework based on field-conditioned
content-style pretraining. The proposed framework first learns control-
lable field-to-field translation across all available field strengths by dis-
entangling anatomical content from field-dependent contrast character-
istics. The pretrained backbone is then adapted to task-specific target
domains. Our model comprises a 3D content encoder, a 3D style en-
coder, a field-conditioned style generator, an AdaIN-modulated decoder,
and a multi-field discriminator. Adversarial learning encourages realistic
target-field appearance, while cycle-consistency, identity, content, style,
and diversity constraints promote anatomical fidelity and controllable
translation. We evaluate the proposed method on MRIxFields data span-
ningfivefieldstrengthsandthreeMRImodalities.Experimentsonpaired
test data demonstrate that the framework can adapt to the three chal-
lenge settings while preserving three-dimensional anatomical structure in
the synthesized volumes. The implementation code is publicly available
at https://github.com/Idea89560041/3D-MRI-Field-Translation.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/MRIxFields2026_001.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=2M2fJiZiV8
BibTex
@InProceedings{PanHao_TaskAdaptive_MICCAISAT2026,
author = { Pang, Haowen AND Hao, Yingqi AND Zhu, Pengli},
title = { { Task-Adaptive 3D Cross-Field MRI Translation via Field-Conditioned Content-Style Pretraining } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17274},
month = {pending},
page = {pending}
}