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
|
Voges I, Raimondi F, McMahon CJ, Ait-Ali L, Babu-Narayan SV, Botnar RM, Burkhardt B, Gabbert DD, Grosse-Wortmann L, Hasan H, Hansmann G, Helbing WA, Krupickova S, Latus H, Martini N, Martins D, Muthurangu V, Ojala T, van Ooij P, Pushparajah K, Rodriguez-Palomares J, Sarikouch S, Grotenhuis HB, Greil FG. Clinical impact of novel CMR technology on patients with congenital heart disease. A scientific statement of the Association for European Pediatric and Congenital Cardiology (AEPC) and the European Association of Cardiovascular Imaging (EACVI) of the ESC. Eur Heart J Cardiovasc Imaging 2024:jeae172. [PMID: 38985851 DOI: 10.1093/ehjci/jeae172] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/28/2024] [Accepted: 07/01/2024] [Indexed: 07/12/2024] Open
Abstract
Cardiovascular magnetic resonance (CMR) imaging is recommended in patients with congenital heart disease (CHD) in clinical practice guidelines as the imaging standard for a large variety of diseases. As CMR is evolving, novel techniques are becoming available. Some of them are already used clinically, whereas others still need further evaluation. In this statement the authors give an overview of relevant new CMR techniques for the assessment of CHD. Studies with reference values for these new techniques are listed in the supplement.
Collapse
Affiliation(s)
- Inga Voges
- Department of Congenital Heart Disease and Pediatric Cardiology, University Hospital Schleswig-Holstein, Campus Kiel, Germany
- German Center for Cardiovascular Research (DZHK), partner site Hamburg/Lübeck/Kiel, Germany
| | | | - Colin J McMahon
- Department of Paediatric Cardiology, Children's Health Ireland at Crumlin, Dublin 12, Ireland
| | - Lamia Ait-Ali
- Institute of clinical Physiology CNR, Massa, Italy
- Heart Hospital, G. Monastery foundation, Massa, Italy
| | - Sonya V Babu-Narayan
- Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
| | - René M Botnar
- School of Biomedical Engineering and Imaging Sciences, King's College London, St. Thomas' Hospital, London, UK
- Institute for Biological and Medical Engineering and School of Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile
| | - Barbara Burkhardt
- Pediatric Heart Center, University Children's Hospital Zurich, Zurich, Switzerland
| | - Dominik D Gabbert
- Department of Congenital Heart Disease and Pediatric Cardiology, University Hospital Schleswig-Holstein, Campus Kiel, Germany
- German Center for Cardiovascular Research (DZHK), partner site Hamburg/Lübeck/Kiel, Germany
| | - Lars Grosse-Wortmann
- Division of Cardiology, Oregon Health and Science University Hospital, Portland, Oregon, United States
| | - Hosan Hasan
- Department of Pediatric Cardiology and Critical Care, Hannover Medical School, Hannover, Germany
- European Pediatric Pulmonary Vascular Disease Network, Berlin, Germany
| | - Georg Hansmann
- Department of Pediatric Cardiology and Critical Care, Hannover Medical School, Hannover, Germany
- European Pediatric Pulmonary Vascular Disease Network, Berlin, Germany
| | - Willem A Helbing
- Department of Pediatrics, division of cardiology, and department of Radiology, Erasmus MC-Sophia children's hospital, Rotterdam, the Netherlands
| | - Sylvia Krupickova
- Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
- Department of Paediatric Cardiology, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, UK
| | - Heiner Latus
- Clinic for Pediatric Cardiology and Congenital Heart Disease Klinikum Stuttgart Germany
| | - Nicola Martini
- Department of Radiology, Fondazione G. Monasterio CNR-Regione Toscana, Pisa, Italy
- U.O.C. Bioingegneria, Fondazione G. Monasterio CNR-Regione Toscana, Pisa, Italy
| | - Duarte Martins
- Pediatric Cardiology Department, Hospital de Santa Cruz, Centro Hospitalar Lisboa Ocidental, Lisbon, Portugal
| | - Vivek Muthurangu
- Centre for Translational Cardiovascular Imaging, Institute of Cardiovascular Science, University College London, London, United Kingdom
| | - Tiina Ojala
- New Children's Hospital Pediatric Research Center, Helsinki University Hospital, Helsinki, Finland
| | - Pim van Ooij
- Department of Radiology & Nuclear Medicine, Amsterdam University Medical Center, location AMC, Amsterdam, the Netherlands
- Department of Pediatric Cardiology, Wilhelmina Children's Hospital/University Medical Center Utrecht, Utrecht, the Netherlands
| | - Kuberan Pushparajah
- School of Biomedical Engineering and Imaging Sciences, King's College London, St. Thomas' Hospital, London, UK
- Department of Congenital Heart Disease, Evelina London Children's Hospital, Westminster Bridge Road, London, SE1 7EH, UK
| | - Jose Rodriguez-Palomares
- CIBER Cardiovascular, Instituto de Salud Carlos III, Madrid, Spain
- European Reference Network for Rare and Low Prevalence Complex Diseases of the Heart, Amsterdam, the Netherlands
- Servicio de Cardiología, Hospital Universitario Vall Hebrón. Institut de Recerca Vall Hebrón (VHIR). Departamento de Medicina, Universitat Autònoma de Barcelona. Barcelona. Spain
| | - Samir Sarikouch
- Department for Cardiothoracic, Transplant, and Vascular Surgery, Hannover Medical School, Hannover, Germany
| | - Heynric B Grotenhuis
- Department of Pediatric Cardiology, Wilhelmina Children's Hospital/University Medical Center Utrecht, Utrecht, the Netherlands
| | - F Gerald Greil
- Department of Pediatrics, UT Southwestern/Children's Health, 1935 Medical District Drive B3.09, Dallas, TX 75235
| |
Collapse
|
3
|
Yang H, Hong K, Baraboo JJ, Fan L, Larsen A, Markl M, Robinson JD, Rigsby CK, Kim D. GRASP reconstruction amplified with view-sharing and KWIC filtering reduces underestimation of peak velocity in highly-accelerated real-time phase-contrast MRI: A preliminary evaluation in pediatric patients with congenital heart disease. Magn Reson Med 2024; 91:1965-1977. [PMID: 38084397 PMCID: PMC10950531 DOI: 10.1002/mrm.29974] [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: 08/10/2023] [Revised: 10/27/2023] [Accepted: 11/27/2023] [Indexed: 02/01/2024]
Abstract
PURPOSE To develop a highly-accelerated, real-time phase contrast (rtPC) MRI pulse sequence with 40 fps frame rate (25 ms effective temporal resolution). METHODS Highly-accelerated golden-angle radial sparse parallel (GRASP) with over regularization may result in temporal blurring, which in turn causes underestimation of peak velocity. Thus, we amplified GRASP performance by synergistically combining view-sharing (VS) and k-space weighted image contrast (KWIC) filtering. In 17 pediatric patients with congenital heart disease (CHD), the conventional GRASP and the proposed GRASP amplified by VS and KWIC (or GRASP + VS + KWIC) reconstruction for rtPC MRI were compared with respect to clinical standard PC MRI in measuring hemodynamic parameters (peak velocity, forward volume, backward volume, regurgitant fraction) at four locations (aortic valve, pulmonary valve, left and right pulmonary arteries). RESULTS The proposed reconstruction method (GRASP + VS + KWIC) achieved better effective spatial resolution (i.e., image sharpness) compared with conventional GRASP, ultimately reducing the underestimation of peak velocity from 17.4% to 6.4%. The hemodynamic metrics (peak velocity, volumes) were not significantly (p > 0.99) different between GRASP + VS + KWIC and clinical PC, whereas peak velocity was significantly (p < 0.007) lower for conventional GRASP. RtPC with GRASP + VS + KWIC also showed the ability to assess beat-to-beat variation and detect the highest peak among peaks. CONCLUSION The synergistic combination of GRASP, VS, and KWIC achieves 25 ms effective temporal resolution (40 fps frame rate), while minimizing the underestimation of peak velocity compared with conventional GRASP.
Collapse
Affiliation(s)
- Huili Yang
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, Illinois, USA
| | - KyungPyo Hong
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
| | - Justin J Baraboo
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, Illinois, USA
| | - Lexiaozi Fan
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, Illinois, USA
| | - Andrine Larsen
- Department of Biomedical Engineering, Lehigh University, Bethlehem, Pennsylvania, USA
| | - Michael Markl
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, Illinois, USA
| | - Joshua D Robinson
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Division of Cardiology, Ann & Robert H. Lurie Children's Hospital, Chicago, Illinois, USA
| | - Cynthia K Rigsby
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Medical Imaging, Ann & Robert H. Lurie Children's Hospital, Chicago, Illinois, USA
| | - Daniel Kim
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
- Department of Biomedical Engineering, McCormick School of Engineering, Northwestern University, Evanston, Illinois, USA
| |
Collapse
|
4
|
Morales MA, Manning WJ, Nezafat R. Present and Future Innovations in AI and Cardiac MRI. Radiology 2024; 310:e231269. [PMID: 38193835 PMCID: PMC10831479 DOI: 10.1148/radiol.231269] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/17/2023] [Revised: 10/21/2023] [Accepted: 10/26/2023] [Indexed: 01/10/2024]
Abstract
Cardiac MRI is used to diagnose and treat patients with a multitude of cardiovascular diseases. Despite the growth of clinical cardiac MRI, complicated image prescriptions and long acquisition protocols limit the specialty and restrain its impact on the practice of medicine. Artificial intelligence (AI)-the ability to mimic human intelligence in learning and performing tasks-will impact nearly all aspects of MRI. Deep learning (DL) primarily uses an artificial neural network to learn a specific task from example data sets. Self-driving scanners are increasingly available, where AI automatically controls cardiac image prescriptions. These scanners offer faster image collection with higher spatial and temporal resolution, eliminating the need for cardiac triggering or breath holding. In the future, fully automated inline image analysis will most likely provide all contour drawings and initial measurements to the reader. Advanced analysis using radiomic or DL features may provide new insights and information not typically extracted in the current analysis workflow. AI may further help integrate these features with clinical, genetic, wearable-device, and "omics" data to improve patient outcomes. This article presents an overview of AI and its application in cardiac MRI, including in image acquisition, reconstruction, and processing, and opportunities for more personalized cardiovascular care through extraction of novel imaging markers.
Collapse
Affiliation(s)
- Manuel A. Morales
- From the Department of Medicine, Cardiovascular Division (M.A.M.,
W.J.M., R.N.), and Department of Radiology (W.J.M.), Beth Israel Deaconess
Medical Center and Harvard Medical School, 330 Brookline Ave, Boston, MA
02215
| | - Warren J. Manning
- From the Department of Medicine, Cardiovascular Division (M.A.M.,
W.J.M., R.N.), and Department of Radiology (W.J.M.), Beth Israel Deaconess
Medical Center and Harvard Medical School, 330 Brookline Ave, Boston, MA
02215
| | - Reza Nezafat
- From the Department of Medicine, Cardiovascular Division (M.A.M.,
W.J.M., R.N.), and Department of Radiology (W.J.M.), Beth Israel Deaconess
Medical Center and Harvard Medical School, 330 Brookline Ave, Boston, MA
02215
| |
Collapse
|
5
|
Moscatelli S, Gatehouse P, Krupickova S, Mohiaddin R, Voges I, Giese D, Nielles-Vallespin S, Pennell DJ. Impact of compressed sensing (CS) acceleration of two-dimensional (2D) flow sequences in clinical paediatric cardiovascular magnetic resonance (CMR). MAGMA (NEW YORK, N.Y.) 2023; 36:869-876. [PMID: 37202654 PMCID: PMC10667407 DOI: 10.1007/s10334-023-01098-8] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/09/2023] [Revised: 05/02/2023] [Accepted: 05/03/2023] [Indexed: 05/20/2023]
Abstract
OBJECTIVES Two-dimensional (2D) through-plane phase-contrast (PC) cine flow imaging assesses shunts and valve regurgitations in paediatric CMR and is considered the reference standard for Clinical quantification of blood Flow (COF). However, longer breath-holds (BH) can reduce compliance with possibly large respiratory manoeuvres altering flow. We hypothesize that reduced BH time by application of CS (Short BH quantification of Flow) (SBOF) retains accuracy while enabling faster, potentially more reliable flows. We investigate the variance between COF and SBOF cine flows. METHODS Main pulmonary artery (MPA) and sinotubular junction (STJ) planes were acquired at 1.5 T in paediatric patients by COF and SBOF. RESULTS 21 patients (mean age 13.9, 10-17y) were enrolled. The BH times were COF mean 11.7 s (range 8.4-20.9 s) vs SBOF mean 6.5 s (min 3.6-9.1 s). The differences and 95% CI between the COF and SBOF flows were LVSV -1.43 ± 13.6(ml/beat), LVCO 0.16 ± 1.35(l/min) and RVSV 2.95 ± 12.3(ml/beat), RVCO 0.27 ± 0.96(l/min), QP/QS were SV 0.04 ± 0.19, CO 0.02 ± 0.23. Variability between COF and SBOF did not exceed intrasession variation of COF. CONCLUSION SBOF reduces breath-hold duration to 56% of COF. RV flow by SBOF was biased compared to COF. The variation (95% CI) between COF and SBOF was similar to the COF intrasession test-retest 95% CI.
Collapse
Affiliation(s)
- Sara Moscatelli
- Department of Paediatric Cardiology, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, UK
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
| | - Peter Gatehouse
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK.
- National Heart and Lung Institute, Imperial College, London, England.
| | - Sylvia Krupickova
- Department of Paediatric Cardiology, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, UK
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
| | - Raad Mohiaddin
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
| | - Inga Voges
- Department of Congenital Heart Disease and Pediatric Cardiology, University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany
| | - Daniel Giese
- Magnetic Resonance, Siemens Healthcare GmbH, Erlangen, Germany
| | - Sonia Nielles-Vallespin
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
| | - Dudley J Pennell
- Department of CMR, Royal Brompton Hospital, Part of Guy's and St Thomas' NHS Foundation Trust, Sydney Street, London, SW3 6NP, UK
- National Heart and Lung Institute, Imperial College, London, England
| |
Collapse
|
6
|
Oscanoa JA, Middione MJ, Alkan C, Yurt M, Loecher M, Vasanawala SS, Ennis DB. Deep Learning-Based Reconstruction for Cardiac MRI: A Review. Bioengineering (Basel) 2023; 10:334. [PMID: 36978725 PMCID: PMC10044915 DOI: 10.3390/bioengineering10030334] [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: 02/02/2023] [Revised: 03/03/2023] [Accepted: 03/03/2023] [Indexed: 03/09/2023] Open
Abstract
Cardiac magnetic resonance (CMR) is an essential clinical tool for the assessment of cardiovascular disease. Deep learning (DL) has recently revolutionized the field through image reconstruction techniques that allow unprecedented data undersampling rates. These fast acquisitions have the potential to considerably impact the diagnosis and treatment of cardiovascular disease. Herein, we provide a comprehensive review of DL-based reconstruction methods for CMR. We place special emphasis on state-of-the-art unrolled networks, which are heavily based on a conventional image reconstruction framework. We review the main DL-based methods and connect them to the relevant conventional reconstruction theory. Next, we review several methods developed to tackle specific challenges that arise from the characteristics of CMR data. Then, we focus on DL-based methods developed for specific CMR applications, including flow imaging, late gadolinium enhancement, and quantitative tissue characterization. Finally, we discuss the pitfalls and future outlook of DL-based reconstructions in CMR, focusing on the robustness, interpretability, clinical deployment, and potential for new methods.
Collapse
Affiliation(s)
- Julio A. Oscanoa
- Department of Bioengineering, Stanford University, Stanford, CA 94305, USA
- Department of Radiology, Stanford University, Stanford, CA 94305, USA
| | | | - Cagan Alkan
- Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA
| | - Mahmut Yurt
- Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA
| | - Michael Loecher
- Department of Radiology, Stanford University, Stanford, CA 94305, USA
| | | | - Daniel B. Ennis
- Department of Radiology, Stanford University, Stanford, CA 94305, USA
| |
Collapse
|