发表机构
Indian Institute of Technology Madras (IITM); Healthcare Technology Innovation Centre; GE John F Welch Technology Center（印度理工学院马德拉斯分校; 医疗技术创新中心; 通用电气约翰·F·韦尔奇技术中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对基于注意力机制的MRI重建模型在高频表示和多模态数据处理上的不足，提出含频谱滤波CNN和动态高通内核生成变压器的网络进行可扩展MRI重建，在未见场景下有良好重建效果，提升了PSNR和SSIM。
AI 中文摘要
注意力机制（AM）可选择性地聚焦于成像任务的关键信息，并捕捉远处像素邻域之间的关系以计算特征表示。加速MRI重建受益于AM，但其更擅长捕捉低频信息，对高频表示能力有限，限制了模型进行平滑重建。此外，基于AM的模型需要针对多模态MRI数据进行特定模式的再训练。为应对这些挑战，我们提出了一种基于神经调节的判别性多光谱AM，用于可扩展的MRI重建，它能传播上下文感知的高频细节以进行高质量重建，并能在多模态MRI的不同未见域中捕捉可重复使用的特征。该网络由一个频谱滤波CNN和一个专注于高频细节的动态高通内核生成变压器组成。我们在监督和自监督学习、基于扩散模型的训练、异构MRI数据下的闭集和开集泛化以及基于解释的分析等比较研究中评估了我们的模型。我们的方法在未见场景下提供了可扩展的高质量重建，PSNR有~1 dB的最佳提升幅度，SSIM有~0.01的提升幅度。
英文摘要
Attention Mechanism (AM) selectively focuses on essential information for imaging tasks and captures relationships between distant pixel neighborhoods to compute feature representations. Accelerated MRI reconstruction benefits from AM, as the imaging process involves Fourier domain measurements that influence image representation non-locally. However, AM-based models are more adept at capturing low-frequency information with limited capacity for high-frequency representations, restricting models to smooth reconstruction. Additionally, AM-based models need mode-specific retraining for multimodal MRI data, as their knowledge is restricted to local contextual variations that may be inadequate to capture transferable features across heterogeneous domains. To address these challenges, we propose a neuromodulation-based discriminative multi-spectral AM for scalable MRI reconstruction that can (i) propagate context-aware high-frequency details for high-quality reconstruction, and (ii) capture features reusable across deviated unseen domains in multimodal MRI. The proposed network consists of a spectral filtering CNN to capture mode-specific transferable features and a dynamic high-pass kernel generation transformer focusing on high-frequency details. We evaluate our model on comparative studies in supervised and self-supervised learning, diffusion model-based training, closed-set and open-set generalization under heterogeneous MRI data, and interpretation-based analysis. Our method offers scalable, high-quality reconstruction with best improvement margins of ~1 dB in PSNR and ~0.01 in SSIM under unseen scenarios. Code: https://github.com/sriprabhar/SHFormer
CommentsPublished in Neural Networks (Elsevier), Vol. 187, Article 107334, July 2025. DOI: 10.1016/j.neunet.2025.107334
Journal refNeural Networks, Vol. 187, Article 107334, July 2025
DOI：10.1016/j.neunet.2025.107334
