🤖 AI Summary
Conventional accelerated MRI reconstruction methods neglect the intrinsic complex-valued structure of k-space data, relying instead on magnitude images or separate real/imaginary channel inputs. Method: This paper introduces AFTNet—the first end-to-end frequency-domain reconstruction framework—operating directly on raw complex-valued k-space measurements. It employs complex-valued neural networks (CVNNs) to model cross-domain mappings between k-space and image domains, integrated with manifold learning for joint feature disentanglement and co-optimization in both domains. Contribution/Results: AFTNet establishes the first native complex-valued, purely frequency-domain reconstruction paradigm, eliminating reliance on image-domain priors or post-processing. Experiments demonstrate significant performance gains over state-of-the-art methods across multiple acceleration factors, with strong generalizability to MRS denoising and multi-contrast reconstruction. The code is publicly available, facilitating preclinical imaging and spectroscopic analysis applications.
📝 Abstract
Deep complex-valued neural networks (CVNNs) provide a powerful way to leverage complex number operations and representations and have succeeded in several phase-based applications. However, previous networks have not fully explored the impact of complex-valued networks in the frequency domain. Here, we introduce a unified complex-valued deep learning framework-Artificial Fourier Transform Network (AFTNet)-which combines domain-manifold learning and CVNNs. AFTNet can be readily used to solve image inverse problems in domain transformation, especially for accelerated magnetic resonance imaging (MRI) reconstruction and other applications. While conventional methods typically utilize magnitude images or treat the real and imaginary components of k-space data as separate channels, our approach directly processes raw k-space data in the frequency domain, utilizing complex-valued operations. This allows for a mapping between the frequency (k-space) and image domain to be determined through cross-domain learning. We show that AFTNet achieves superior accelerated MRI reconstruction compared to existing approaches. Furthermore, our approach can be applied to various tasks, such as denoised magnetic resonance spectroscopy (MRS) reconstruction and datasets with various contrasts. The AFTNet presented here is a valuable preprocessing component for different preclinical studies and provides an innovative alternative for solving inverse problems in imaging and spectroscopy. The code is available at: https://github.com/yanting-yang/AFT-Net.
