UniMo: Universal Motion Correction For Medical Images without Network Retraining

2024-09-21

arXiv.org

AI Summary

Existing medical image motion correction methods exhibit strong modality dependence, limiting generalizability to unseen modalities and yielding insufficient accuracy for large-range motion (e.g., fetal imaging). This paper proposes UniMo, a universal motion correction framework that achieves zero-shot, cross-modal motion artifact correction—spanning MRI, ultrasound, CT, and X-ray—using a single training run, without iterative inference or modality-specific retraining. Key innovations include: (1) a modality-agnostic motion correction paradigm; (2) equivariant filters coupled with geometric deformation enhancers to improve robustness in modeling local deformations; and (3) a joint shape–image learning framework to strengthen cross-modal generalization. Evaluated across four modalities, UniMo significantly outperforms state-of-the-art methods, especially under large-displacement scenarios, where correction accuracy is markedly improved. The implementation is publicly available.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

In this paper, we introduce a Universal Motion Correction (UniMo) framework, leveraging deep neural networks to tackle the challenges of motion correction across diverse imaging modalities. Our approach employs advanced neural network architectures with equivariant filters, overcoming the limitations of current models that require iterative inference or retraining for new image modalities. UniMo enables one-time training on a single modality while maintaining high stability and adaptability for inference across multiple unseen image modalities. We developed a joint learning framework that integrates multimodal knowledge from both shape and images that faithfully improve motion correction accuracy despite image appearance variations. UniMo features a geometric deformation augmenter that enhances the robustness of global motion correction by addressing any local deformations whether they are caused by object deformations or geometric distortions, and also generates augmented data to improve the training process. Our experimental results, conducted on various datasets with four different image modalities, demonstrate that UniMo surpasses existing motion correction methods in terms of accuracy. By offering a comprehensive solution to motion correction, UniMo marks a significant advancement in medical imaging, especially in challenging applications with wide ranges of motion, such as fetal imaging. The code for this work is available online, https://github.com/IntelligentImaging/UNIMO/.

Problem

• Medical Image Correction
• Adaptive Learning
• Fetal Imaging

Innovation

• UniMo
• Deep Learning
• Motion Correction