DuDoCAF: Dual-Domain Cross-Attention Fusion with Recurrent Transformer for Fast Multi-contrast MR Imaging

Lyu, Jun; Sui, Bin; Wang, Chengyan; Tian, Yapeng; Dou, Qi; Qin, Jing

doi:10.1007/978-3-031-16446-0_45

Cited by 19 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Performance evaluation. We compare our methods with other baseline deep learning methods in three conventions of MRI reconstruction: image-domain [24,34,6,9,13], dual-domain [24,34,33] and reference-protocol-guided dual-domain reconstruction [29,6,34,33,19]. All reference-guided methods are self-implemented besides Du-DoRNet and examined without considering multi-modal fusion modules for controlled backbone comparisons.…”

Section: Settings and Resultsmentioning

confidence: 99%

“…We further unite these methods considering their backbones. [6,29] adopt Dense-Unet; [33,19,13] share similar Swin-Transformer backbones derived from SwinIR [16]. [33] proposed k-space filling using the reference protocol for de-aliasing initially in self-supervised reconstruction, yet it does not improve the model's performance when fully-supervised.…”

Section: Settings and Resultsmentioning

confidence: 99%

“…where F is 2D discrete Fourier Transform and the approximation function P x (•; θ x ) is used to predict a reconstructed signal x r given its parameter θ x and any undersampled input. A few works [34,5,19] leverage the relationship between image and k-space domain using Fourier Transform pairs (F, F −1 ) and transform MRI reconstruction into a multivariable optimization problem described as…”

Section: Undersampled Mri Reconstructionmentioning

confidence: 99%

“…Recently, a growing number of works [21,29,6,34,33,19] utilize a fast-to-acquire fully-sampled auxiliary MRI protocol to guide the reconstruction of a slow protocol.…”

Section: Undersampled Mri Reconstructionmentioning

confidence: 99%

“…Its Multihead Self-Attention (MSA) mechanism captures long-range interactions among contexts globally or inside local windows [17]. As the success of Transformer is now indisputable in computer vision, Transformer has shown great potential for undersampled MRI reconstruction as well [13,33,8,7,19]. Yet performant, ViTs have not fully substituted CNNs as ViTs require a larger amount of training data due to a low inductive bias and have longer training schedules [30,32].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

DuDoRNeXt: A hybrid model for dual-domain undersampled MRI reconstruction

Gao¹,

Zhou²

2023

Preprint

View full text Add to dashboard Cite

Undersampled MRI reconstruction is crucial for accelerating clinical scanning procedures. Recent deep learning methods for MRI reconstruction adopt CNN or ViT as backbone, which lack in utilizing the complementary properties of CNN and ViT. In this paper, we propose DuDoRNeXt, whose backbone hybridizes CNN and ViT in an domainspecific, intra-stage way. Besides our hybrid vertical layout design, we introduce domain-specific modules for dual-domain reconstruction, namely image-domain parallel local detail enhancement and k-space global initialization. We evaluate different conventions of MRI reconstruction including image-domain, k-space-domain, and dual-domain reconstruction with a reference protocol on the IXI dataset and an in-house multicontrast dataset. DuDoRNeXt achieves significant improvements over competing deep learning methods.

show abstract

Section: Settings and Resultsmentioning

confidence: 99%

Section: Settings and Resultsmentioning

confidence: 99%

Section: Undersampled Mri Reconstructionmentioning

confidence: 99%

“…Recently, a growing number of works [21,29,6,34,33,19] utilize a fast-to-acquire fully-sampled auxiliary MRI protocol to guide the reconstruction of a slow protocol.…”

Section: Undersampled Mri Reconstructionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

DuDoRNeXt: A hybrid model for dual-domain undersampled MRI reconstruction

Gao¹,

Zhou²

2023

Preprint

View full text Add to dashboard Cite

show abstract

Comprehensive review of Transformer‐based models in neuroscience, neurology, and psychiatry

Cong,

Wang,

Zhou

et al. 2024

Brain-X

View full text Add to dashboard Cite

This comprehensive review aims to clarify the growing impact of Transformer‐based models in the fields of neuroscience, neurology, and psychiatry. Originally developed as a solution for analyzing sequential data, the Transformer architecture has evolved to effectively capture complex spatiotemporal relationships and long‐range dependencies that are common in biomedical data. Its adaptability and effectiveness in deciphering intricate patterns within medical studies have established it as a key tool in advancing our understanding of neural functions and disorders, representing a significant departure from traditional computational methods. The review begins by introducing the structure and principles of Transformer architectures. It then explores their applicability, ranging from disease diagnosis and prognosis to the evaluation of cognitive processes and neural decoding. The specific design modifications tailored for these applications and their subsequent impact on performance are also discussed. We conclude by providing a comprehensive assessment of recent advancements, prevailing challenges, and future directions, highlighting the shift in neuroscientific research and clinical practice towards an artificial intelligence‐centric paradigm, particularly given the prominence of Transformer architecture in the most successful large pre‐trained models. This review serves as an informative reference for researchers, clinicians, and professionals who are interested in understanding and harnessing the transformative potential of Transformer‐based models in neuroscience, neurology, and psychiatry.

show abstract