Leveraging Image Editing Foundation Models for Data-Efficient CT Metal Artifact Reduction

TL;DR

This paper introduces a data-efficient, foundation model-based approach for CT metal artifact reduction, leveraging visual priors and multi-reference conditioning to outperform existing methods with minimal training data.

Contribution

It adapts a general-purpose vision-language diffusion model with parameter-efficient LoRA for effective, data-efficient CT artifact reduction, emphasizing domain adaptation and multi-reference conditioning.

Findings

Achieves state-of-the-art performance on the AAPM CT-MAR benchmark.

Reduces training data requirements by two orders of magnitude.

Effectively mitigates hallucinations through domain adaptation.

Abstract

Metal artifacts from high-attenuation implants severely degrade CT image quality, obscuring critical anatomical structures and posing a challenge for standard deep learning methods that require extensive paired training data. We propose a paradigm shift: reframing artifact reduction as an in-context reasoning task by adapting a general-purpose vision-language diffusion foundation model via parameter-efficient Low-Rank Adaptation (LoRA). By leveraging rich visual priors, our approach achieves effective artifact suppression with only 16 to 128 paired training examples reducing data requirements by two orders of magnitude. Crucially, we demonstrate that domain adaptation is essential for hallucination mitigation; without it, foundation models interpret streak artifacts as erroneous natural objects (e.g., waffles or petri dishes). To ground the restoration, we propose a multi-reference conditioning strategy where clean anatomical exemplars from unrelated subjects are provided alongside the corrupted input, enabling the model to exploit category-specific context to infer uncorrupted anatomy. Extensive evaluation on the AAPM CT-MAR benchmark demonstrates that our method achieves state-of-the-art performance on perceptual and radiological-feature metrics . This work establishes that foundation models, when appropriately adapted, offer a scalable alternative for interpretable, data-efficient medical image reconstruction. Code is available at https://github.com/ahmetemirdagi/CT-EditMAR.