Deep Learning with Domain Adaptation for Accelerated Projection Reconstruction MR

By Yoseob Han, Jaejun Yoo, Jong Chul Ye. 2017

245 citations recorded by the source corpus.

Purpose: The radial k-space trajectory is a well-established sampling trajectory used in conjunction with magnetic resonance imaging. However, the radial k-space trajectory requires a large number of radial lines for high-resolution reconstruction. Increasing the number of radial lines causes longer acquisition time, making it more difficult for routine clinical use. On the other hand, if we reduce the number of radial lines, streaking artifact patterns are unavoidable. To solve this problem, we propose a novel deep learning approach with domain adaptation to restore high-resolution MR images from under-sampled k-space data. Methods: The proposed deep network removes the streaking artifacts from the artifact corrupted images. To address the situation given the limited available data, we propose a domain adaptation scheme that employs a pre-trained network using a large number of x-ray computed tomography (CT) or synthesized radial MR datasets, which is then fine-tuned with only a few radial MR datasets. Results: The proposed method outperforms existing compressed sensing algorithms, such as the total variation and PR-FOCUSS methods. In addition, the calculation time is several orders of magnitude faster than the total variation and PR-FOCUSS methods.Moreover, we found that pre-training using CT or MR data from similar organ data is more important than pre-training using data from the same modality for different organ. Conclusion: We demonstrate the possibility of a domain-adaptation when only a limited amount of MR data is available. The proposed method surpasses the existing compressed sensing algorithms in terms of the image quality and computation time.

Subfields: Biomedical Engineering, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

An overview of deep learning in medical imaging focusing on MRI (2018, neutral): “[141, 142, 143, 144], a fundamental field rapidly progressing.”
Deep Residual Learning for Accelerated MRI Using Magnitude and Phase Networks (2018, background): “One of the most widely used deep learning network architecture for inverse problems is U-net [24], [25], which incorporates bypass connection and pooling/unpooling layers.”
Domain Adaptation for Medical Image Analysis: A Survey (2021, methodology): “Similar strategy is also used in [91] where a network is pre-trained using a large number of X-ray computed tomography (CT) and synthesized radial MRI datasets and then fine-tuned with only a few labeled target MRI scans.”
Super-Resolution Musculoskeletal MRI Using Deep Learning (2018, methodology): “30 After the DeepResolve network is trained, the estimated residual f ð X il Þ is calculated directly and a high-resolution super-resolution image f X isr”
Scan-specific Robust Artificial-neural-networks for k-space Interpolation (RAKI) Reconstruction: Database-free Deep Learning for Fast Imaging (2018, methodology): “Big data training has led to different regularization approaches for MR images, most often as regularizers without an explicit formula and based on CNNs, 19,22,23,43 or as unsupervised learning of parameters for variational networks.”
Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks (2020, unknown): “For example, based on the similarity between projection MR and CT, Han et al. proposed to train the network by using CT data and then adapting the network parameter for MR reconstruction with fine-tuning [18].”
CNN-Based Projected Gradient Descent for Consistent CT Image Reconstruction (2017, methodology): “Rather than reconstructing the image from the measurements y directly, the most successful strategies have been to train the CNN as a regressor between a rough initial reconstruction Ay, where A : RM → RN , and the final, desired reconstruction [16], [17].”