FIND-Net -- Fourier-Integrated Network with Dictionary Kernels for Metal Artifact Reduction

2025-08-14

AI Summary

CT metal artifacts severely degrade image quality, compromising clinical diagnosis and treatment planning. To address this, we propose a novel frequency-domain–spatial-domain collaborative deep learning framework. Our method innovatively integrates fast Fourier convolution (FFC) with learnable Gaussian filtering and introduces a dictionary kernel-driven frequency-domain modeling mechanism to achieve precise artifact suppression while preserving anatomical fidelity. Evaluated on both synthetic and real clinical CT datasets, the proposed approach significantly outperforms state-of-the-art methods: it reduces mean absolute error (MAE) by 3.07%, improves structural similarity index (SSIM) by 0.18, and increases peak signal-to-noise ratio (PSNR) by 0.90 dB. The framework demonstrates superior artifact reduction capability without sacrificing diagnostic utility, thereby enhancing both quantitative performance and clinical applicability.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Metal artifacts, caused by high-density metallic implants in computed tomography (CT) imaging, severely degrade image quality, complicating diagnosis and treatment planning. While existing deep learning algorithms have achieved notable success in Metal Artifact Reduction (MAR), they often struggle to suppress artifacts while preserving structural details. To address this challenge, we propose FIND-Net (Fourier-Integrated Network with Dictionary Kernels), a novel MAR framework that integrates frequency and spatial domain processing to achieve superior artifact suppression and structural preservation. FIND-Net incorporates Fast Fourier Convolution (FFC) layers and trainable Gaussian filtering, treating MAR as a hybrid task operating in both spatial and frequency domains. This approach enhances global contextual understanding and frequency selectivity, effectively reducing artifacts while maintaining anatomical structures. Experiments on synthetic datasets show that FIND-Net achieves statistically significant improvements over state-of-the-art MAR methods, with a 3.07% MAE reduction, 0.18% SSIM increase, and 0.90% PSNR improvement, confirming robustness across varying artifact complexities. Furthermore, evaluations on real-world clinical CT scans confirm FIND-Net's ability to minimize modifications to clean anatomical regions while effectively suppressing metal-induced distortions. These findings highlight FIND-Net's potential for advancing MAR performance, offering superior structural preservation and improved clinical applicability. Code is available at https://github.com/Farid-Tasharofi/FIND-Net

Problem

• Reducing metal artifacts in CT imaging caused by implants
• Preserving structural details while suppressing artifacts
• Integrating frequency and spatial domain processing for MAR

Innovation

• Integrates frequency and spatial domain processing
• Uses Fast Fourier Convolution and Gaussian filtering
• Enhances artifact suppression and structural preservation