MR Image Reconstruction via Zero-Shot Learned Generative Adversarial Transformers

Salman Ul Hassan Dar, Yılmaz Korkmaz, Tolga ÇukurProceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition 2023 · DOI

Abstract

Deep neural network models have demonstrated state-of-the-art performance in MR image reconstruction. These models require information regarding imaging operators during training, which limits their generalization. A recent framework is based on zero-shot learned generative models that learn MR image priors during training and couple imaging operator during inference on test acquisitions. Such models are however based on convolutional architectures that suffer from sub-optimal capture of long range dependencies. Here, we propose a novel architecture based on zero-shot learned generative adversarial transformers that enables efficient capture of long range dependencies via cross-attention transformers while removing reliance on imaging operator during training.
BibTeX
@inproceedings{dar2023mr,
  author = {Salman Ul Hassan Dar and Yılmaz Korkmaz and Tolga Çukur},
  title = {{MR Image Reconstruction via Zero-Shot Learned Generative Adversarial Transformers}},
  booktitle = {Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition},
  year = {2023},
  doi = {10.58530/2022/0502},
}

Loading related content...

Figure