1
|
Dawood P, Breuer F, Stebani J, Burd P, Homolya I, Oberberger J, Jakob PM, Blaimer M. Iterative training of robust k-space interpolation networks for improved image reconstruction with limited scan specific training samples. Magn Reson Med 2023; 89:812-827. [PMID: 36226661 DOI: 10.1002/mrm.29482] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/18/2022] [Revised: 09/12/2022] [Accepted: 09/20/2022] [Indexed: 12/13/2022]
Abstract
PURPOSE To evaluate an iterative learning approach for enhanced performance of robust artificial-neural-networks for k-space interpolation (RAKI), when only a limited amount of training data (auto-calibration signals [ACS]) are available for accelerated standard 2D imaging. METHODS In a first step, the RAKI model was tailored for the case of limited training data amount. In the iterative learning approach (termed iterative RAKI [iRAKI]), the tailored RAKI model is initially trained using original and augmented ACS obtained from a linear parallel imaging reconstruction. Subsequently, the RAKI convolution filters are refined iteratively using original and augmented ACS extracted from the previous RAKI reconstruction. Evaluation was carried out on 200 retrospectively undersampled in vivo datasets from the fastMRI neuro database with different contrast settings. RESULTS For limited training data (18 and 22 ACS lines for R = 4 and R = 5, respectively), iRAKI outperforms standard RAKI by reducing residual artifacts and yields better noise suppression when compared to standard parallel imaging, underlined by quantitative reconstruction quality metrics. Additionally, iRAKI shows better performance than both GRAPPA and standard RAKI in case of pre-scan calibration with varying contrast between training- and undersampled data. CONCLUSION RAKI benefits from the iterative learning approach, which preserves the noise suppression feature, but requires less original training data for the accurate reconstruction of standard 2D images thereby improving net acceleration.
Collapse
Affiliation(s)
- Peter Dawood
- Department of Physics, University of Würzburg, Würzburg, Germany
| | - Felix Breuer
- Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany
| | - Jannik Stebani
- Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany
| | - Paul Burd
- Institute for Theoretical Physics and Astrophysics, University of Würzburg, Würzburg, Germany
| | - István Homolya
- Brain Imaging Centre, Research Centre for Natural Sciences, Budapest, Hungary
| | - Johannes Oberberger
- Department of Internal Medicine I, University Hospital Würzburg, Würzburg, Germany
| | - Peter M Jakob
- Department of Physics, University of Würzburg, Würzburg, Germany
| | - Martin Blaimer
- Magnetic Resonance and X-ray Imaging Department, Fraunhofer Institute for Integrated Circuits IIS, Division Development Center X-Ray Technology, Würzburg, Germany
| |
Collapse
|
2
|
Zhang C, Moeller S, Demirel OB, Uğurbil K, Akçakaya M. Residual RAKI: A hybrid linear and non-linear approach for scan-specific k-space deep learning. Neuroimage 2022; 256:119248. [PMID: 35487456 PMCID: PMC9179026 DOI: 10.1016/j.neuroimage.2022.119248] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2021] [Revised: 04/07/2022] [Accepted: 04/23/2022] [Indexed: 10/31/2022] Open
Abstract
Parallel imaging is the most clinically used acceleration technique for magnetic resonance imaging (MRI) in part due to its easy inclusion into routine acquisitions. In k-space based parallel imaging reconstruction, sub-sampled k-space data are interpolated using linear convolutions. At high acceleration rates these methods have inherent noise amplification and reduced image quality. On the other hand, non-linear deep learning methods provide improved image quality at high acceleration, but the availability of training databases for different scans, as well as their interpretability hinder their adaptation. In this work, we present an extension of Robust Artificial-neural-networks for k-space Interpolation (RAKI), called residual-RAKI (rRAKI), which achieves scan-specific machine learning reconstruction using a hybrid linear and non-linear methodology. In rRAKI, non-linear CNNs are trained jointly with a linear convolution implemented via a skip connection. In effect, the linear part provides a baseline reconstruction, while the non-linear CNN that runs in parallel provides further reduction of artifacts and noise arising from the linear part. The explicit split between the linear and non-linear aspects of the reconstruction also help improve interpretability compared to purely non-linear methods. Experiments were conducted on the publicly available fastMRI datasets, as well as high-resolution anatomical imaging, comparing GRAPPA and its variants, compressed sensing, RAKI, Scan Specific Artifact Reduction in K-space (SPARK) and the proposed rRAKI. Additionally, highly-accelerated simultaneous multi-slice (SMS) functional MRI reconstructions were also performed, where the proposed rRAKI was compred to Read-out SENSE-GRAPPA and RAKI. Our results show that the proposed rRAKI method substantially improves the image quality compared to conventional parallel imaging, and offers sharper images compared to SPARK and ℓ1-SPIRiT. Furthermore, rRAKI shows improved preservation of time-varying dynamics compared to both parallel imaging and RAKI in highly-accelerated SMS fMRI.
Collapse
Affiliation(s)
- Chi Zhang
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, USA; Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN 55455, USA
| | - Steen Moeller
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN 55455, USA
| | - Omer Burak Demirel
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, USA; Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN 55455, USA
| | - Kâmil Uğurbil
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN 55455, USA
| | - Mehmet Akçakaya
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, USA; Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN 55455, USA.
| |
Collapse
|
3
|
Takeshima H. Deep Learning and Its Application to Function Approximation for MR in Medicine: An Overview. Magn Reson Med Sci 2021; 21:553-568. [PMID: 34544924 DOI: 10.2463/mrms.rev.2021-0040] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2022] Open
Abstract
This article presents an overview of deep learning (DL) and its applications to function approximation for MR in medicine. The aim of this article is to help readers develop various applications of DL. DL has made a large impact on the literature of many medical sciences, including MR. However, its technical details are not easily understandable for non-experts of machine learning (ML).The first part of this article presents an overview of DL and its related technologies, such as artificial intelligence (AI) and ML. AI is explained as a function that can receive many inputs and produce many outputs. ML is a process of fitting the function to training data. DL is a kind of ML, which uses a composite of many functions to approximate the function of interest. This composite function is called a deep neural network (DNN), and the functions composited into a DNN are called layers. This first part also covers the underlying technologies required for DL, such as loss functions, optimization, initialization, linear layers, non-linearities, normalization, recurrent neural networks, regularization, data augmentation, residual connections, autoencoders, generative adversarial networks, model and data sizes, and complex-valued neural networks.The second part of this article presents an overview of the applications of DL in MR and explains how functions represented as DNNs are applied to various applications, such as RF pulse, pulse sequence, reconstruction, motion correction, spectroscopy, parameter mapping, image synthesis, and segmentation.
Collapse
Affiliation(s)
- Hidenori Takeshima
- Advanced Technology Research Department, Research and Development Center, Canon Medical Systems Corporation
| |
Collapse
|
4
|
Risk BB, Murden RJ, Wu J, Nebel MB, Venkataraman A, Zhang Z, Qiu D. Which multiband factor should you choose for your resting-state fMRI study? Neuroimage 2021; 234:117965. [PMID: 33744454 PMCID: PMC8159874 DOI: 10.1016/j.neuroimage.2021.117965] [Citation(s) in RCA: 26] [Impact Index Per Article: 8.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/15/2021] [Revised: 03/05/2021] [Accepted: 03/06/2021] [Indexed: 12/30/2022] Open
Abstract
Multiband acquisition, also called simultaneous multislice, has become a popular technique in resting-state functional connectivity studies. Multiband (MB) acceleration leads to a higher temporal resolution but also leads to spatially heterogeneous noise amplification, suggesting the costs may be greater in areas such as the subcortex. We evaluate MB factors of 2, 3, 4, 6, 8, 9, and 12 with 2 mm isotropic voxels, and additionally 2 mm and 3.3 mm single-band acquisitions, on a 32-channel head coil. Noise amplification was greater in deeper brain regions, including subcortical regions. Correlations were attenuated by noise amplification, which resulted in spatially varying biases that were more severe at higher MB factors. Temporal filtering decreased spatial biases in correlations due to noise amplification, but also tended to decrease effect sizes. In seed-based correlation maps, left-right putamen connectivity and thalamo-motor connectivity were highest in the single-band 3.3 mm protocol. In correlation matrices, MB 4, 6, and 8 had a greater number of significant correlations than the other acquisitions (both with and without temporal filtering). We recommend single-band 3.3 mm for seed-based subcortical analyses, and MB 4 provides a reasonable balance for studies analyzing both seed-based correlation maps and connectivity matrices. In multiband studies including secondary analyses of large-scale datasets, we recommend reporting effect sizes or test statistics instead of correlations. If correlations are reported, temporal filtering (or another method for thermal noise removal) should be used. The Emory Multiband Dataset is available on OpenNeuro.
Collapse
Affiliation(s)
- Benjamin B Risk
- Department of Biostatistics and Bioinformatics, Atlanta, GA, United States.
| | - Raphiel J Murden
- Department of Biostatistics and Bioinformatics, Atlanta, GA, United States
| | - Junjie Wu
- Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States
| | - Mary Beth Nebel
- Center for Neurodevelopmental and Imaging Research, Kennedy Krieger Institute, Baltimore, MD, United States; Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, United States
| | - Arun Venkataraman
- Department of Physics and Astronomy, University of Rochester, Rochester, NY, United States
| | - Zhengwu Zhang
- Department of Statistics and Operations Research, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States
| | - Deqiang Qiu
- Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States
| |
Collapse
|