1
|
Wang S, Wu R, Jia S, Diakite A, Li C, Liu Q, Zheng H, Ying L. Knowledge-driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un-supervised learning. Magn Reson Med 2024; 92:496-518. [PMID: 38624162 DOI: 10.1002/mrm.30105] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/03/2023] [Revised: 03/19/2024] [Accepted: 03/20/2024] [Indexed: 04/17/2024]
Abstract
Deep learning (DL) has emerged as a leading approach in accelerating MRI. It employs deep neural networks to extract knowledge from available datasets and then applies the trained networks to reconstruct accurate images from limited measurements. Unlike natural image restoration problems, MRI involves physics-based imaging processes, unique data properties, and diverse imaging tasks. This domain knowledge needs to be integrated with data-driven approaches. Our review will introduce the significant challenges faced by such knowledge-driven DL approaches in the context of fast MRI along with several notable solutions, which include learning neural networks and addressing different imaging application scenarios. The traits and trends of these techniques have also been given which have shifted from supervised learning to semi-supervised learning, and finally, to unsupervised learning methods. In addition, MR vendors' choices of DL reconstruction have been provided along with some discussions on open questions and future directions, which are critical for the reliable imaging systems.
Collapse
Affiliation(s)
- Shanshan Wang
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
| | - Ruoyou Wu
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
| | - Sen Jia
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
| | - Alou Diakite
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
- University of Chinese Academy of Sciences, Beijing, China
| | - Cheng Li
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
| | - Qiegen Liu
- Department of Electronic Information Engineering, Nanchang University, Nanchang, China
| | - Hairong Zheng
- Paul C Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
| | - Leslie Ying
- Department of Biomedical Engineering and Department of Electrical Engineering, The State University of New York, Buffalo, New York, USA
| |
Collapse
|
2
|
Bian W, Jang A, Liu F. Improving quantitative MRI using self-supervised deep learning with model reinforcement: Demonstration for rapid T1 mapping. Magn Reson Med 2024; 92:98-111. [PMID: 38342980 PMCID: PMC11055673 DOI: 10.1002/mrm.30045] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/04/2023] [Revised: 12/21/2023] [Accepted: 01/23/2024] [Indexed: 02/13/2024]
Abstract
PURPOSE This paper proposes a novel self-supervised learning framework that uses model reinforcement, REference-free LAtent map eXtraction with MOdel REinforcement (RELAX-MORE), for accelerated quantitative MRI (qMRI) reconstruction. The proposed method uses an optimization algorithm to unroll an iterative model-based qMRI reconstruction into a deep learning framework, enabling accelerated MR parameter maps that are highly accurate and robust. METHODS Unlike conventional deep learning methods which require large amounts of training data, RELAX-MORE is a subject-specific method that can be trained on single-subject data through self-supervised learning, making it accessible and practically applicable to many qMRI studies. Using quantitativeT 1 $$ {\mathrm{T}}_1 $$ mapping as an example, the proposed method was applied to the brain, knee and phantom data. RESULTS The proposed method generates high-quality MR parameter maps that correct for image artifacts, removes noise, and recovers image features in regions of imperfect image conditions. Compared with other state-of-the-art conventional and deep learning methods, RELAX-MORE significantly improves efficiency, accuracy, robustness, and generalizability for rapid MR parameter mapping. CONCLUSION This work demonstrates the feasibility of a new self-supervised learning method for rapid MR parameter mapping, that is readily adaptable to the clinical translation of qMRI.
Collapse
Affiliation(s)
- Wanyu Bian
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, Massachusetts, USA
- Harvard Medical School, Boston, Massachusetts, USA
| | - Albert Jang
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, Massachusetts, USA
- Harvard Medical School, Boston, Massachusetts, USA
| | - Fang Liu
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, Massachusetts, USA
- Harvard Medical School, Boston, Massachusetts, USA
| |
Collapse
|
3
|
Heydari A, Ahmadi A, Kim TH, Bilgic B. Joint MAPLE: Accelerated joint T 1 and T 2 * $$ {{\mathrm{T}}_2}^{\ast } $$ mapping with scan-specific self-supervised networks. Magn Reson Med 2024; 91:2294-2309. [PMID: 38181183 PMCID: PMC11007829 DOI: 10.1002/mrm.29989] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/19/2023] [Revised: 10/30/2023] [Accepted: 12/11/2023] [Indexed: 01/07/2024]
Abstract
PURPOSE Quantitative MRI finds important applications in clinical and research studies. However, it is encoding intensive and may suffer from prohibitively long scan times. Accelerated MR parameter mapping techniques have been developed to help address these challenges. Here, an accelerated joint T1,T 2 * $$ {{\mathrm{T}}_2}^{\ast } $$ , frequency and proton density mapping technique with scan-specific self-supervised network reconstruction is proposed to synergistically combine parallel imaging, model-based, and deep learning approaches to speed up parameter mapping. METHODS Proposed framework, Joint MAPLE, includes parallel imaging, signal modeling, and data consistency blocks which are optimized jointly in a combined loss function. A scan-specific self-supervised reconstruction is embedded into the framework, which takes advantage of multi-contrast data from a multi-echo, multi-flip angle, gradient echo acquisition. RESULTS In comparison with parallel reconstruction techniques powered by low-rank methods, emerging scan specific networks, and model-basedT 2 * $$ {{\mathrm{T}}_2}^{\ast } $$ estimation approaches, the proposed framework reduces the reconstruction error in parameter maps by approximately two-fold on average at acceleration rates as high as R = 16 with uniform sampling. It can outperform evaluated parallel reconstruction techniques up to four-fold on average in the presence of challenging sub-sampling masks. It is observed that Joint MAPLE performs well at extreme acceleration rates of R = 25 and R = 36 with error values less than 20%. CONCLUSION Joint MAPLE enables higher fidelity parameter estimation at high acceleration rates by synergistically combining parallel imaging and model-based parameter mapping and exploiting multi-echo, multi-flip angle datasets. Utilizing a scan-specific self-supervised reconstruction obviates the need for large data sets for training while improving the parameter estimation ability.
Collapse
Affiliation(s)
- Amir Heydari
- Department of Industrial Engineering and Management Systems, Amirkabir University of Technology, Tehran, Iran
| | - Abbas Ahmadi
- Department of Industrial Engineering and Management Systems, Amirkabir University of Technology, Tehran, Iran
| | - Tae Hyung Kim
- Department of Computer Engineering, Hongik University, Seoul, Korea
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, United States
- Radiology, Harvard Medical School, Boston, MA, United States
| | - Berkin Bilgic
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, United States
- Radiology, Harvard Medical School, Boston, MA, United States
- Harvard/MIT Health Sciences and Technology, Cambridge, MA, United States
| |
Collapse
|
4
|
Lu Q, Li J, Lian Z, Zhang X, Feng Q, Chen W, Ma J, Feng Y. A model-based MR parameter mapping network robust to substantial variations in acquisition settings. Med Image Anal 2024; 94:103148. [PMID: 38554550 DOI: 10.1016/j.media.2024.103148] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/20/2022] [Revised: 12/03/2023] [Accepted: 03/20/2024] [Indexed: 04/01/2024]
Abstract
Deep learning methods show great potential for the efficient and precise estimation of quantitative parameter maps from multiple magnetic resonance (MR) images. Current deep learning-based MR parameter mapping (MPM) methods are mostly trained and tested using data with specific acquisition settings. However, scan protocols usually vary with centers, scanners, and studies in practice. Thus, deep learning methods applicable to MPM with varying acquisition settings are highly required but still rarely investigated. In this work, we develop a model-based deep network termed MMPM-Net for robust MPM with varying acquisition settings. A deep learning-based denoiser is introduced to construct the regularization term in the nonlinear inversion problem of MPM. The alternating direction method of multipliers is used to solve the optimization problem and then unrolled to construct MMPM-Net. The variation in acquisition parameters can be addressed by the data fidelity component in MMPM-Net. Extensive experiments are performed on R2 mapping and R1 mapping datasets with substantial variations in acquisition settings, and the results demonstrate that the proposed MMPM-Net method outperforms other state-of-the-art MR parameter mapping methods both qualitatively and quantitatively.
Collapse
Affiliation(s)
- Qiqi Lu
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China; Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-Inspired Intelligence & Key Laboratory of Mental Health of the Ministry of Education & Guangdong-Hong Kong Joint Laboratory for Psychiatric Disorders, Southern Medical University, Guangzhou 510000, China; Department of Radiology, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde, Foshan), Foshan 528000, China
| | - Jialong Li
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China; Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-Inspired Intelligence & Key Laboratory of Mental Health of the Ministry of Education & Guangdong-Hong Kong Joint Laboratory for Psychiatric Disorders, Southern Medical University, Guangzhou 510000, China; Department of Radiology, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde, Foshan), Foshan 528000, China
| | - Zifeng Lian
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China; Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-Inspired Intelligence & Key Laboratory of Mental Health of the Ministry of Education & Guangdong-Hong Kong Joint Laboratory for Psychiatric Disorders, Southern Medical University, Guangzhou 510000, China; Department of Radiology, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde, Foshan), Foshan 528000, China
| | - Xinyuan Zhang
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China
| | - Qianjin Feng
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China
| | - Wufan Chen
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China
| | - Jianhua Ma
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China
| | - Yanqiu Feng
- School of Biomedical Engineering, Southern Medical University, Guangzhou 510000, China; Guangdong Provincial Key Laboratory of Medical Image Processing & Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou 510000, China; Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-Inspired Intelligence & Key Laboratory of Mental Health of the Ministry of Education & Guangdong-Hong Kong Joint Laboratory for Psychiatric Disorders, Southern Medical University, Guangzhou 510000, China; Department of Radiology, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde, Foshan), Foshan 528000, China.
| |
Collapse
|
5
|
Han W, Wang Y, Li T, Dong Y, Dang Y, He L, Xu L, Zhou Y, Li Y, Wang X. A CT-based integrated model for preoperative prediction of occult lymph node metastasis in early tongue cancer. PeerJ 2024; 12:e17254. [PMID: 38685941 PMCID: PMC11057426 DOI: 10.7717/peerj.17254] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/29/2023] [Accepted: 03/26/2024] [Indexed: 05/02/2024] Open
Abstract
Background Occult lymph node metastasis (OLNM) is an essential prognostic factor for early-stage tongue cancer (cT1-2N0M0) and a determinant of treatment decisions. Therefore, accurate prediction of OLNM can significantly impact the clinical management and outcomes of patients with tongue cancer. The aim of this study was to develop and validate a multiomics-based model to predict OLNM in patients with early-stage tongue cancer. Methods The data of 125 patients diagnosed with early-stage tongue cancer (cT1-2N0M0) who underwent primary surgical treatment and elective neck dissection were retrospectively analyzed. A total of 100 patients were randomly assigned to the training set and 25 to the test set. The preoperative contrast-enhanced computed tomography (CT) and clinical data on these patients were collected. Radiomics features were extracted from the primary tumor as the region of interest (ROI) on CT images, and correlation analysis and the least absolute shrinkage and selection operator (LASSO) method were used to identify the most relevant features. A support vector machine (SVM) classifier was constructed and compared with other machine learning algorithms. With the same method, a clinical model was built and the peri-tumoral and intra-tumoral images were selected as the input for the deep learning model. The stacking ensemble technique was used to combine the multiple models. The predictive performance of the integrated model was evaluated for accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC-ROC), and compared with expert assessment. Internal validation was performed using a stratified five-fold cross-validation approach. Results Of the 125 patients, 41 (32.8%) showed OLNM on postoperative pathological examination. The integrated model achieved higher predictive performance compared with the individual models, with an accuracy of 84%, a sensitivity of 100%, a specificity of 76.5%, and an AUC-ROC of 0.949 (95% CI [0.870-1.000]). In addition, the performance of the integrated model surpassed that of younger doctors and was comparable to the evaluation of experienced doctors. Conclusions The multiomics-based model can accurately predict OLNM in patients with early-stage tongue cancer, and may serve as a valuable decision-making tool to determine the appropriate treatment and avoid unnecessary neck surgery in patients without OLNM.
Collapse
Affiliation(s)
- Wei Han
- Department of Maxillofacial and Otorhinolaryngological Oncology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Tianjin Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Yingshu Wang
- Department of Radiology, the Affiliated Cancer Hospital of Zhengzhou University, Zhengzhou, China
| | - Tao Li
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Yuke Dong
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Yanwei Dang
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Liang He
- Department of Maxillofacial and Otorhinolaryngological Oncology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Tianjin Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| | - Lianfang Xu
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Yuhao Zhou
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Yujie Li
- Department of Otolaryngology, Head and Neck Surgery, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China
| | - Xudong Wang
- Department of Maxillofacial and Otorhinolaryngological Oncology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Tianjin Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China
| |
Collapse
|
6
|
Altmann S, Abello Mercado MA, Brockstedt L, Kronfeld A, Clifford B, Feiweier T, Uphaus T, Groppa S, Brockmann MA, Othman AE. Ultrafast Brain MRI Protocol at 1.5 T Using Deep Learning and Multi-shot EPI. Acad Radiol 2023; 30:2988-2998. [PMID: 37211480 DOI: 10.1016/j.acra.2023.04.019] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2023] [Revised: 04/14/2023] [Accepted: 04/17/2023] [Indexed: 05/23/2023]
Abstract
RATIONALE AND OBJECTIVES To evaluate clinical feasibility and image quality of a comprehensive ultrafast brain MRI protocol with multi-shot echo planar imaging and deep learning-enhanced reconstruction at 1.5T. MATERIALS AND METHODS Thirty consecutive patients who underwent clinically indicated MRI at a 1.5 T scanner were prospectively included. A conventional MRI (c-MRI) protocol, including T1-, T2-, T2*-, T2-FLAIR, and diffusion-weighted images (DWI)-weighted sequences were acquired. In addition, ultrafast brain imaging with deep learning-enhanced reconstruction and multi-shot EPI (DLe-MRI) was performed. Subjective image quality was evaluated by three readers using a 4-point Likert scale. To assess interrater agreement, Fleiss' kappa (ϰ) was determined. For objective image analysis, relative signal intensity levels for grey matter, white matter, and cerebrospinal fluid were calculated. RESULTS Time of acquisition (TA) of c-MRI protocols added up to 13:55 minutes, whereas the TA of DLe-MRI-based protocol added up to 3:04 minutes, resulting in a time reduction of 78%. All DLe-MRI acquisitions yielded diagnostic image quality with good absolute values for subjective image quality. C-MRI demonstrated slight advantages for DWI in overall subjective image quality (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.87 [+/- 0.37], P = .04) and diagnostic confidence (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.83 [+/- 3.83], P = .01). For most evaluated quality scores, moderate interobserver agreement was found. Objective image evaluation revealed comparable results for both techniques. CONCLUSION DLe-MRI is feasible and allows for highly accelerated comprehensive brain MRI within 3minutes at 1.5 T with good image quality. This technique may potentially strengthen the role of MRI in neurological emergencies.
Collapse
Affiliation(s)
- Sebastian Altmann
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.).
| | - Mario Alberto Abello Mercado
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.)
| | - Lavinia Brockstedt
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.)
| | - Andrea Kronfeld
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.)
| | - Bryan Clifford
- Siemens Medical Solutions USA, Boston, Massachusetts (B.C.)
| | | | - Timo Uphaus
- Department of Neurology, University Medical Center Mainz, Johannes Gutenberg University, Mainz, Germany (T.U., S.G.)
| | - Sergiu Groppa
- Department of Neurology, University Medical Center Mainz, Johannes Gutenberg University, Mainz, Germany (T.U., S.G.)
| | - Marc A Brockmann
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.)
| | - Ahmed E Othman
- Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckst. 1, 55131 Mainz, Germany (S.A., M.A.M., L.B., A.K., M.A.B., A.E.O.)
| |
Collapse
|
7
|
Hellström M, Löfstedt T, Garpebring A. Denoising and uncertainty estimation in parameter mapping with approximate Bayesian deep image priors. Magn Reson Med 2023; 90:2557-2571. [PMID: 37582257 DOI: 10.1002/mrm.29823] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/26/2022] [Revised: 06/26/2023] [Accepted: 07/18/2023] [Indexed: 08/17/2023]
Abstract
PURPOSE To mitigate the problem of noisy parameter maps with high uncertainties by casting parameter mapping as a denoising task based on Deep Image Priors. METHODS We extend the concept of denoising with Deep Image Prior (DIP) into parameter mapping by treating the output of an image-generating network as a parametrization of tissue parameter maps. The method implicitly denoises the parameter mapping process by filtering low-level image features with an untrained convolutional neural network (CNN). Our implementation includes uncertainty estimation from Bernoulli approximate variational inference, implemented with MC dropout, which provides model uncertainty in each voxel of the denoised parameter maps. The method is modular, so the specifics of different applications (e.g., T1 mapping) separate into application-specific signal equation blocks. We evaluate the method on variable flip angle T1 mapping, multi-echo T2 mapping, and apparent diffusion coefficient mapping. RESULTS We found that deep image prior adapts successfully to several applications in parameter mapping. In all evaluations, the method produces noise-reduced parameter maps with decreased uncertainty compared to conventional methods. The downsides of the proposed method are the long computational time and the introduction of some bias from the denoising prior. CONCLUSION DIP successfully denoise the parameter mapping process and applies to several applications with limited hyperparameter tuning. Further, it is easy to implement since DIP methods do not use network training data. Although time-consuming, uncertainty information from MC dropout makes the method more robust and provides useful information when properly calibrated.
Collapse
Affiliation(s)
- Max Hellström
- Department of Radiation Sciences, Umeå University, Umeå, Sweden
| | - Tommy Löfstedt
- Department of Radiation Sciences, Umeå University, Umeå, Sweden
- Department of Computing Science, Umeå University, Umeå, Sweden
| | | |
Collapse
|
8
|
Singh D, Monga A, de Moura HL, Zhang X, Zibetti MVW, Regatte RR. Emerging Trends in Fast MRI Using Deep-Learning Reconstruction on Undersampled k-Space Data: A Systematic Review. Bioengineering (Basel) 2023; 10:1012. [PMID: 37760114 PMCID: PMC10525988 DOI: 10.3390/bioengineering10091012] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/31/2023] [Revised: 08/22/2023] [Accepted: 08/24/2023] [Indexed: 09/29/2023] Open
Abstract
Magnetic Resonance Imaging (MRI) is an essential medical imaging modality that provides excellent soft-tissue contrast and high-resolution images of the human body, allowing us to understand detailed information on morphology, structural integrity, and physiologic processes. However, MRI exams usually require lengthy acquisition times. Methods such as parallel MRI and Compressive Sensing (CS) have significantly reduced the MRI acquisition time by acquiring less data through undersampling k-space. The state-of-the-art of fast MRI has recently been redefined by integrating Deep Learning (DL) models with these undersampled approaches. This Systematic Literature Review (SLR) comprehensively analyzes deep MRI reconstruction models, emphasizing the key elements of recently proposed methods and highlighting their strengths and weaknesses. This SLR involves searching and selecting relevant studies from various databases, including Web of Science and Scopus, followed by a rigorous screening and data extraction process using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. It focuses on various techniques, such as residual learning, image representation using encoders and decoders, data-consistency layers, unrolled networks, learned activations, attention modules, plug-and-play priors, diffusion models, and Bayesian methods. This SLR also discusses the use of loss functions and training with adversarial networks to enhance deep MRI reconstruction methods. Moreover, we explore various MRI reconstruction applications, including non-Cartesian reconstruction, super-resolution, dynamic MRI, joint learning of reconstruction with coil sensitivity and sampling, quantitative mapping, and MR fingerprinting. This paper also addresses research questions, provides insights for future directions, and emphasizes robust generalization and artifact handling. Therefore, this SLR serves as a valuable resource for advancing fast MRI, guiding research and development efforts of MRI reconstruction for better image quality and faster data acquisition.
Collapse
Affiliation(s)
- Dilbag Singh
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, NY 10016, USA; (A.M.); (H.L.d.M.); (X.Z.); (M.V.W.Z.)
| | | | | | | | | | - Ravinder R. Regatte
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, NY 10016, USA; (A.M.); (H.L.d.M.); (X.Z.); (M.V.W.Z.)
| |
Collapse
|
9
|
刘 羽, 楚 智, 张 煜. [Physical model-based cascaded generative adversarial networks for accelerating quantitative multi-parametric magnetic resonance imaging]. NAN FANG YI KE DA XUE XUE BAO = JOURNAL OF SOUTHERN MEDICAL UNIVERSITY 2023; 43:1402-1409. [PMID: 37712278 PMCID: PMC10505569 DOI: 10.12122/j.issn.1673-4254.2023.08.18] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Subscribe] [Scholar Register] [Received: 03/15/2023] [Indexed: 09/16/2023]
Abstract
OBJECTIVE To explore the feasibility and interpretation of physical model- based cascaded generative adversarial networks for accelerating quantitative multi-echo multi-parametric magnetic resonance imaging using raw multi-echo multicoil k-space data. METHODS A physical model-based cascaded generative adversarial network is proposed to enhance image feature information to obtain high-quality reconstructed images using joint training of multi-domain information and learning of key parameters required for image reconstruction through a system matrix and adaptively optimizing the k-space generator and image generator structures. Raw multi-echo multi-coil k-space data are used to accelerate multi-contrast multi-parametric magnetic resonance imaging. A physically driven deep learning reconstruction method is used to increase the generalization capability and improve the model performance by building a system matrix function instead of direct end-to-end training of the model. RESULTS In terms of overall image quality, the proposed model achieved significant improvements compared to other methods. On an 80- case test set, the average PSNR value of the reconstructed images was 34.13, SSIM was 0.965, and NRMSE was 0.114. In terms of multi-contrast multi-parametric image reconstruction, the model achieved PSNR values of 38.87 for PDW, 35.62 for T1W, and 34.38 for T2* Map, which were significantly better than those of other methods for quantitative evaluation. The model also produced clearer features of the brain gray matter, white matter, and cerebrospinal fluid. Furthermore, compared with the existing methods with a reconstruction time difference of less than 10%, the proposed method achieved the highest improvement of up to 20% in the metrics of PSNR, SSIM, and NRMSE. CONCLUSION Compared with other existing methods, the physical model-based cascaded generative adversarial networks can reconstruct more image details and features, thus improving the quality and accuracy of the reconstructed images.
Collapse
Affiliation(s)
- 羽轩 刘
- />南方医科大学生物医学工程学院//广东省医学图像处理重点实验室,广东 广州 510515School of Biomedical Engineering, Southern Medical University//Guangdong Provincial Key Laboratory of Medical Image Processing, Guangzhou 510515, China
| | - 智钦 楚
- />南方医科大学生物医学工程学院//广东省医学图像处理重点实验室,广东 广州 510515School of Biomedical Engineering, Southern Medical University//Guangdong Provincial Key Laboratory of Medical Image Processing, Guangzhou 510515, China
| | - 煜 张
- />南方医科大学生物医学工程学院//广东省医学图像处理重点实验室,广东 广州 510515School of Biomedical Engineering, Southern Medical University//Guangdong Provincial Key Laboratory of Medical Image Processing, Guangzhou 510515, China
| |
Collapse
|
10
|
Bian W, Jang A, Liu F. Magnetic Resonance Parameter Mapping using Self-supervised Deep Learning with Model Reinforcement. ARXIV 2023:arXiv:2307.13211v1. [PMID: 37547657 PMCID: PMC10402181] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Subscribe] [Scholar Register] [Indexed: 08/08/2023]
Abstract
This paper proposes a novel self-supervised learning method, RELAX-MORE, for quantitative MRI (qMRI) reconstruction. The proposed method uses an optimization algorithm to unroll a model-based qMRI reconstruction into a deep learning framework, enabling the generation of highly accurate and robust MR parameter maps at imaging acceleration. Unlike conventional deep learning methods requiring a large amount of training data, RELAX-MORE is a subject-specific method that can be trained on single-subject data through self-supervised learning, making it accessible and practically applicable to many qMRI studies. Using the quantitative T 1 mapping as an example at different brain, knee and phantom experiments, the proposed method demonstrates excellent performance in reconstructing MR parameters, correcting imaging artifacts, removing noises, and recovering image features at imperfect imaging conditions. Compared with other state-of-the-art conventional and deep learning methods, RELAX-MORE significantly improves efficiency, accuracy, robustness, and generalizability for rapid MR parameter mapping. This work demonstrates the feasibility of a new self-supervised learning method for rapid MR parameter mapping, with great potential to enhance the clinical translation of qMRI.
Collapse
Affiliation(s)
- Wanyu Bian
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129 USA
| | - Albert Jang
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129 USA
| | - Fang Liu
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129 USA
| |
Collapse
|
11
|
Zibetti MVW, Menon RG, de Moura HL, Zhang X, Kijowski R, Regatte RR. Updates on Compositional MRI Mapping of the Cartilage: Emerging Techniques and Applications. J Magn Reson Imaging 2023; 58:44-60. [PMID: 37010113 PMCID: PMC10323700 DOI: 10.1002/jmri.28689] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/18/2023] [Revised: 03/06/2023] [Accepted: 03/06/2023] [Indexed: 04/04/2023] Open
Abstract
Osteoarthritis (OA) is a widely occurring degenerative joint disease that is severely debilitating and causes significant socioeconomic burdens to society. Magnetic resonance imaging (MRI) is the preferred imaging modality for the morphological evaluation of cartilage due to its excellent soft tissue contrast and high spatial resolution. However, its utilization typically involves subjective qualitative assessment of cartilage. Compositional MRI, which refers to the quantitative characterization of cartilage using a variety of MRI methods, can provide important information regarding underlying compositional and ultrastructural changes that occur during early OA. Cartilage compositional MRI could serve as early imaging biomarkers for the objective evaluation of cartilage and help drive diagnostics, disease characterization, and response to novel therapies. This review will summarize current and ongoing state-of-the-art cartilage compositional MRI techniques and highlight emerging methods for cartilage compositional MRI including MR fingerprinting, compressed sensing, multiexponential relaxometry, improved and robust radio-frequency pulse sequences, and deep learning-based acquisition, reconstruction, and segmentation. The review will also briefly discuss the current challenges and future directions for adopting these emerging cartilage compositional MRI techniques for use in clinical practice and translational OA research studies. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 2.
Collapse
Affiliation(s)
- Marcelo V. W. Zibetti
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Rajiv G. Menon
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Hector L. de Moura
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Xiaoxia Zhang
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Richard Kijowski
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Ravinder R. Regatte
- Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| |
Collapse
|
12
|
Cascade of Denoising and Mapping Neural Networks for MRI R2* Relaxometry of Iron-Loaded Liver. Bioengineering (Basel) 2023; 10:bioengineering10020209. [PMID: 36829703 PMCID: PMC9952355 DOI: 10.3390/bioengineering10020209] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/18/2023] [Revised: 02/01/2023] [Accepted: 02/02/2023] [Indexed: 02/09/2023] Open
Abstract
MRI of effective transverse relaxation rate (R2*) measurement is a reliable method for liver iron concentration quantification. However, R2* mapping can be degraded by noise, especially in the case of iron overload. This study aimed to develop a deep learning method for MRI R2* relaxometry of an iron-loaded liver using a two-stage cascaded neural network. The proposed method, named CadamNet, combines two convolutional neural networks separately designed for image denoising and parameter mapping into a cascade framework, and the physics-based R2* decay model was incorporated in training the mapping network to enforce data consistency further. CadamNet was trained using simulated liver data with Rician noise, which was constructed from clinical liver data. The performance of CadamNet was quantitatively evaluated on simulated data with varying noise levels as well as clinical liver data and compared with the single-stage parameter mapping network (MappingNet) and two conventional model-based R2* mapping methods. CadamNet consistently achieved high-quality R2* maps and outperformed MappingNet at varying noise levels. Compared with conventional R2* mapping methods, CadamNet yielded R2* maps with lower errors, higher quality, and substantially increased efficiency. In conclusion, the proposed CadamNet enables accurate and efficient iron-loaded liver R2* mapping, especially in the presence of severe noise.
Collapse
|
13
|
Free-breathing and instantaneous abdominal T 2 mapping via single-shot multiple overlapping-echo acquisition and deep learning reconstruction. Eur Radiol 2023:10.1007/s00330-023-09417-2. [PMID: 36692597 DOI: 10.1007/s00330-023-09417-2] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/12/2022] [Revised: 10/12/2022] [Accepted: 01/01/2023] [Indexed: 01/25/2023]
Abstract
OBJECTIVES To develop a real-time abdominal T2 mapping method without requiring breath-holding or respiratory-gating. METHODS The single-shot multiple overlapping-echo detachment (MOLED) pulse sequence was employed to achieve free-breathing T2 mapping of the abdomen. Deep learning was used to untangle the non-linear relationship between the MOLED signal and T2 mapping. A synthetic data generation flow based on Bloch simulation, modality synthesis, and randomization was proposed to overcome the inadequacy of real-world training set. RESULTS The results from simulation and in vivo experiments demonstrated that our method could deliver high-quality T2 mapping. The average NMSE and R2 values of linear regression in the digital phantom experiments were 0.0178 and 0.9751. Pearson's correlation coefficient between our predicted T2 and reference T2 in the phantom experiments was 0.9996. In the measurements for the patients, real-time capture of the T2 value changes of various abdominal organs before and after contrast agent injection was realized. A total of 33 focal liver lesions were detected in the group, and the mean and standard deviation of T2 values were 141.1 ± 50.0 ms for benign and 63.3 ± 16.0 ms for malignant lesions. The coefficients of variance in a test-retest experiment were 2.9%, 1.2%, 0.9%, 3.1%, and 1.8% for the liver, kidney, gallbladder, spleen, and skeletal muscle, respectively. CONCLUSIONS Free-breathing abdominal T2 mapping is achieved in about 100 ms on a clinical MRI scanner. The work paved the way for the development of real-time dynamic T2 mapping in the abdomen. KEY POINTS • MOLED achieves free-breathing abdominal T2 mapping in about 100 ms, enabling real-time capture of T2 value changes due to CA injection in abdominal organs. • Synthetic data generation flow mitigates the issue of lack of sizable abdominal training datasets.
Collapse
|
14
|
Tolpadi AA, Han M, Calivà F, Pedoia V, Majumdar S. Region of interest-specific loss functions improve T 2 quantification with ultrafast T 2 mapping MRI sequences in knee, hip and lumbar spine. Sci Rep 2022; 12:22208. [PMID: 36564430 PMCID: PMC9789075 DOI: 10.1038/s41598-022-26266-z] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/22/2022] [Accepted: 12/13/2022] [Indexed: 12/24/2022] Open
Abstract
MRI T2 mapping sequences quantitatively assess tissue health and depict early degenerative changes in musculoskeletal (MSK) tissues like cartilage and intervertebral discs (IVDs) but require long acquisition times. In MSK imaging, small features in cartilage and IVDs are crucial for diagnoses and must be preserved when reconstructing accelerated data. To these ends, we propose region of interest-specific postprocessing of accelerated acquisitions: a recurrent UNet deep learning architecture that provides T2 maps in knee cartilage, hip cartilage, and lumbar spine IVDs from accelerated T2-prepared snapshot gradient-echo acquisitions, optimizing for cartilage and IVD performance with a multi-component loss function that most heavily penalizes errors in those regions. Quantification errors in knee and hip cartilage were under 10% and 9% from acceleration factors R = 2 through 10, respectively, with bias for both under 3 ms for most of R = 2 through 12. In IVDs, mean quantification errors were under 12% from R = 2 through 6. A Gray Level Co-Occurrence Matrix-based scheme showed knee and hip pipelines outperformed state-of-the-art models, retaining smooth textures for most R and sharper ones through moderate R. Our methodology yields robust T2 maps while offering new approaches for optimizing and evaluating reconstruction algorithms to facilitate better preservation of small, clinically relevant features.
Collapse
Affiliation(s)
- Aniket A Tolpadi
- Department of Radiology and Biomedical Imaging, University of California, 1700, 4th Street, San Francisco, CA, 94158, USA.
| | - Misung Han
- Department of Radiology and Biomedical Imaging, University of California, 1700, 4th Street, San Francisco, CA, 94158, USA
| | - Francesco Calivà
- Department of Radiology and Biomedical Imaging, University of California, 1700, 4th Street, San Francisco, CA, 94158, USA
| | - Valentina Pedoia
- Department of Radiology and Biomedical Imaging, University of California, 1700, 4th Street, San Francisco, CA, 94158, USA
| | - Sharmila Majumdar
- Department of Radiology and Biomedical Imaging, University of California, 1700, 4th Street, San Francisco, CA, 94158, USA
| |
Collapse
|
15
|
Velasco C, Fletcher TJ, Botnar RM, Prieto C. Artificial intelligence in cardiac magnetic resonance fingerprinting. Front Cardiovasc Med 2022; 9:1009131. [PMID: 36204566 PMCID: PMC9530662 DOI: 10.3389/fcvm.2022.1009131] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Accepted: 08/30/2022] [Indexed: 11/13/2022] Open
Abstract
Magnetic resonance fingerprinting (MRF) is a fast MRI-based technique that allows for multiparametric quantitative characterization of the tissues of interest in a single acquisition. In particular, it has gained attention in the field of cardiac imaging due to its ability to provide simultaneous and co-registered myocardial T1 and T2 mapping in a single breath-held cardiac MRF scan, in addition to other parameters. Initial results in small healthy subject groups and clinical studies have demonstrated the feasibility and potential of MRF imaging. Ongoing research is being conducted to improve the accuracy, efficiency, and robustness of cardiac MRF. However, these improvements usually increase the complexity of image reconstruction and dictionary generation and introduce the need for sequence optimization. Each of these steps increase the computational demand and processing time of MRF. The latest advances in artificial intelligence (AI), including progress in deep learning and the development of neural networks for MRI, now present an opportunity to efficiently address these issues. Artificial intelligence can be used to optimize candidate sequences and reduce the memory demand and computational time required for reconstruction and post-processing. Recently, proposed machine learning-based approaches have been shown to reduce dictionary generation and reconstruction times by several orders of magnitude. Such applications of AI should help to remove these bottlenecks and speed up cardiac MRF, improving its practical utility and allowing for its potential inclusion in clinical routine. This review aims to summarize the latest developments in artificial intelligence applied to cardiac MRF. Particularly, we focus on the application of machine learning at different steps of the MRF process, such as sequence optimization, dictionary generation and image reconstruction.
Collapse
Affiliation(s)
- Carlos Velasco
- School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom
- *Correspondence: Carlos Velasco
| | - Thomas J. Fletcher
- School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom
| | - René M. Botnar
- School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom
- Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile
- Millennium Institute for Intelligent Healthcare Engineering, Santiago, Chile
| | - Claudia Prieto
- School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom
- Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile
- Millennium Institute for Intelligent Healthcare Engineering, Santiago, Chile
| |
Collapse
|
16
|
Feng L, Ma D, Liu F. Rapid MR relaxometry using deep learning: An overview of current techniques and emerging trends. NMR IN BIOMEDICINE 2022; 35:e4416. [PMID: 33063400 PMCID: PMC8046845 DOI: 10.1002/nbm.4416] [Citation(s) in RCA: 24] [Impact Index Per Article: 12.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/27/2020] [Revised: 08/25/2020] [Accepted: 09/09/2020] [Indexed: 05/08/2023]
Abstract
Quantitative mapping of MR tissue parameters such as the spin-lattice relaxation time (T1 ), the spin-spin relaxation time (T2 ), and the spin-lattice relaxation in the rotating frame (T1ρ ), referred to as MR relaxometry in general, has demonstrated improved assessment in a wide range of clinical applications. Compared with conventional contrast-weighted (eg T1 -, T2 -, or T1ρ -weighted) MRI, MR relaxometry provides increased sensitivity to pathologies and delivers important information that can be more specific to tissue composition and microenvironment. The rise of deep learning in the past several years has been revolutionizing many aspects of MRI research, including image reconstruction, image analysis, and disease diagnosis and prognosis. Although deep learning has also shown great potential for MR relaxometry and quantitative MRI in general, this research direction has been much less explored to date. The goal of this paper is to discuss the applications of deep learning for rapid MR relaxometry and to review emerging deep-learning-based techniques that can be applied to improve MR relaxometry in terms of imaging speed, image quality, and quantification robustness. The paper is comprised of an introduction and four more sections. Section 2 describes a summary of the imaging models of quantitative MR relaxometry. In Section 3, we review existing "classical" methods for accelerating MR relaxometry, including state-of-the-art spatiotemporal acceleration techniques, model-based reconstruction methods, and efficient parameter generation approaches. Section 4 then presents how deep learning can be used to improve MR relaxometry and how it is linked to conventional techniques. The final section concludes the review by discussing the promise and existing challenges of deep learning for rapid MR relaxometry and potential solutions to address these challenges.
Collapse
Affiliation(s)
- Li Feng
- Biomedical Engineering and Imaging Institute and Department of Radiology, Icahn School of Medicine at Mount Sinai, New York, New York
| | - Dan Ma
- Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio
| | - Fang Liu
- Department of Radiology, Massachusetts General Hospital, Harvard University, Boston, Massachusetts
| |
Collapse
|
17
|
So S, Park HW, Kim B, Fritz FJ, Poser BA, Roebroeck A, Bilgic B. BUDA-MESMERISE: Rapid acquisition and unsupervised parameter estimation for T 1 , T 2 , M 0 , B 0 , and B 1 maps. Magn Reson Med 2022; 88:292-308. [PMID: 35344611 DOI: 10.1002/mrm.29228] [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: 08/12/2021] [Revised: 02/21/2022] [Accepted: 02/22/2022] [Indexed: 11/10/2022]
Abstract
PURPOSE Rapid acquisition scheme and parameter estimation method are proposed to acquire distortion-free spin- and stimulated-echo signals and combine the signals with a physics-driven unsupervised network to estimate T1 , T2 , and proton density (M0 ) parameter maps, along with B0 and B1 information from the acquired signals. THEORY AND METHODS An imaging sequence with three 90° RF pulses is utilized to acquire spin- and stimulated-echo signals. We utilize blip-up/-down acquisition to eliminate geometric distortion incurred by the effects of B0 inhomogeneity on rapid EPI acquisitions. For multislice imaging, echo-shifting is applied to utilize dead time between the second and third RF pulses to encode information from additional slice positions. To estimate parameter maps from the spin- and stimulated-echo signals with high fidelity, 2 estimation methods, analytic fitting and a novel unsupervised deep neural network method, are developed. RESULTS The proposed acquisition provided distortion-free T1 , T2 , relative proton density (M0), B0 , and B1 maps with high fidelity both in phantom and in vivo brain experiments. From the rapidly acquired spin- and stimulated-echo signals, analytic fitting and the network-based method were able to estimate T1 , T2 , M0 , B0 , and B1 maps with high accuracy. Network estimates demonstrated noise robustness owing to the fact that the convolutional layers take information into account from spatially adjacent voxels. CONCLUSION The proposed acquisition/reconstruction technique enabled whole-brain acquisition of coregistered, distortion-free, T1 , T2 , M0 , B0 , and B1 maps at 1 × 1 × 5 mm3 resolution in 50 s. The proposed unsupervised neural network provided noise-robust parameter estimates from this rapid acquisition.
Collapse
Affiliation(s)
- Seohee So
- School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
| | - Hyun Wook Park
- School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
| | - Byungjai Kim
- School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
| | - Francisco J Fritz
- Institute of Systems Neuroscience, Center for Experimental Medicine, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany
| | - Benedikt A Poser
- Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands
| | - Alard Roebroeck
- Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands
| | - Berkin Bilgic
- Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, Massachusetts, USA.,Department of Radiology, Harvard Medical School, Charlestown, Massachusetts, USA.,Harvard-MIT Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
| |
Collapse
|
18
|
Liu F, Kijowski R, El Fakhri G, Feng L. Magnetic resonance parameter mapping using model-guided self-supervised deep learning. Magn Reson Med 2021; 85:3211-3226. [PMID: 33464652 PMCID: PMC9185837 DOI: 10.1002/mrm.28659] [Citation(s) in RCA: 32] [Impact Index Per Article: 10.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2020] [Revised: 11/15/2020] [Accepted: 12/07/2020] [Indexed: 12/25/2022]
Abstract
PURPOSE To develop a model-guided self-supervised deep learning MRI reconstruction framework called reference-free latent map extraction (RELAX) for rapid quantitative MR parameter mapping. METHODS Two physical models are incorporated for network training in RELAX, including the inherent MR imaging model and a quantitative model that is used to fit parameters in quantitative MRI. By enforcing these physical model constraints, RELAX eliminates the need for full sampled reference data sets that are required in standard supervised learning. Meanwhile, RELAX also enables direct reconstruction of corresponding MR parameter maps from undersampled k-space. Generic sparsity constraints used in conventional iterative reconstruction, such as the total variation constraint, can be additionally included in the RELAX framework to improve reconstruction quality. The performance of RELAX was tested for accelerated T1 and T2 mapping in both simulated and actually acquired MRI data sets and was compared with supervised learning and conventional constrained reconstruction for suppressing noise and/or undersampling-induced artifacts. RESULTS In the simulated data sets, RELAX generated good T1 /T2 maps in the presence of noise and/or undersampling artifacts, comparable to artifact/noise-free ground truth. The inclusion of a spatial total variation constraint helps improve image quality. For the in vivo T1 /T2 mapping data sets, RELAX achieved superior reconstruction quality compared with conventional iterative reconstruction, and similar reconstruction performance to supervised deep learning reconstruction. CONCLUSION This work has demonstrated the initial feasibility of rapid quantitative MR parameter mapping based on self-supervised deep learning. The RELAX framework may also be further extended to other quantitative MRI applications by incorporating corresponding quantitative imaging models.
Collapse
Affiliation(s)
- Fang Liu
- Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
| | - Richard Kijowski
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin, USA
| | - Georges El Fakhri
- Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
| | - Li Feng
- Biomedical Engineering and Imaging Institute and Department of Radiology, Icahn School of Medicine at Mount Sinai, New York, USA
| |
Collapse
|