Systematic Review and Meta-analysis of AI-driven MRI Motion Artifact Detection and Correction

2025-09-05

AI Summary

This study addresses key bottlenecks in MRI motion artifact correction—namely, poor generalizability, reliance on paired training data, and propensity for perceptual distortion. We systematically review and conduct a meta-analysis of deep learning–based generative models (e.g., GANs and autoencoders) for artifact detection and correction. Leveraging a multicenter dataset, we train models and perform cross-institutional validation, demonstrating statistically significant improvements in quantitative and qualitative image quality over conventional methods. However, our analysis reveals critical limitations in out-of-distribution generalization and unsupervised/weakly supervised adaptation. To bridge these gaps, we propose a standardized framework for constructing benchmark datasets tailored to MRI artifact tasks and introduce a reproducibility-aware reporting protocol. These contributions provide both methodological foundations and practical guidelines for robust deployment of generative AI in medical imaging quality control.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Background: To systematically review and perform a meta-analysis of artificial intelligence (AI)-driven methods for detecting and correcting magnetic resonance imaging (MRI) motion artifacts, assessing current developments, effectiveness, challenges, and future research directions. Methods: A comprehensive systematic review and meta-analysis were conducted, focusing on deep learning (DL) approaches, particularly generative models, for the detection and correction of MRI motion artifacts. Quantitative data were extracted regarding utilized datasets, DL architectures, and performance metrics. Results: DL, particularly generative models, show promise for reducing motion artifacts and improving image quality; however, limited generalizability, reliance on paired training data, and risk of visual distortions remain key challenges that motivate standardized datasets and reporting. Conclusions: AI-driven methods, particularly DL generative models, show significant potential for improving MRI image quality by effectively addressing motion artifacts. However, critical challenges must be addressed, including the need for comprehensive public datasets, standardized reporting protocols for artifact levels, and more advanced, adaptable DL techniques to reduce reliance on extensive paired datasets. Addressing these aspects could substantially enhance MRI diagnostic accuracy, reduce healthcare costs, and improve patient care outcomes.

Problem

• Detecting and correcting MRI motion artifacts using AI
• Assessing effectiveness and challenges of deep learning methods
• Improving MRI image quality and diagnostic accuracy

Innovation

• Deep learning generative models for artifact correction
• Systematic review and meta-analysis methodology
• Addressing limited generalizability with standardized datasets