Self-Navigated Multi-Contrast And Quantitative Whole Heart 3D Magnetic Resonance Imaging

Lead Research Organisation: King's College London
Department Name: Imaging & Biomedical Engineering

Abstract

Cardiovascular disease (CVD) remains the leading single cause of death worldwide despite improvements in prevention and advances in diagnosis and treatment. The two major causes for CVD death are sudden coronary atherothrombosis due to plaque rupture and subsequent thrombus formation and heart failure following myocardial infarction (MI) due to adverse myocardial remodeling and subsequent infarct expansion. Detection of coronary atherosclerosis and prediction of adverse myocardial remodeling remain challenging with current imaging techniques. While x-ray coronary angiography is the gold standard for the detection of coronary stenosis it has limited value for the detection and characterization of coronary plaque. Multislice computed tomography (MSCT) is an excellent non-invasive alternative for the detection of coronary stenosis and has some ability to visualize and characterize coronary plaque but its diagnostic use is limited to patients without coronary calcification. Positron electron tomography (PET) has good diagnostic sensitivity for myocardial perfusion and viability assessment and recently also has been shown to have potential for coronary plaque visualization but suffers from low spatial resolution and radiation exposure and is not widely available. Echocardiography is the clinical gold standard for the assessment of left ventricular function and wall motion abnormalities. It is cheap and easy to use but is heavily operator dependent. Due to the above limitations there is a need for the development of an alternative and non-invasive imaging test that allows for comprehensive cardiac assessment without the above restrictions.

Magnetic resonance imaging (MRI) is considered the gold standard for the assessment of cardiac anatomy, left ventricular (LF) function (CINE-MRI), myocardial viability (LGE-MRI) and perfusion (MR-perfusion) due to its excellent soft tissue contrast, high spatial resolution and lack of ionizing radiation according to a Society for Magnetic Resonance (SCMR) expert consensus statement. Recent clinical research studies also have demonstrated its usefulness for quantitative myocardial tissue characterization (T1 and T2 relaxation time mapping) and its ability to differentiate between healthy and diseased tissue. However, a key limitation of the current MRI acquisition scheme is that all imaging sequences (e.g. CINE, LGE, T1 and T2 mapping, coronary MR angiography (MRA), etc.) are acquired sequentially, in different geometric orientations, at different breath-hold positions or using time inefficient navigator gating methods. This imposes several challenges: (1) radiographers need high expertise to perform the complex examination, (2) patients have to perform multiple (>30) breathholds which can be very challenging in sick patients, (3) the duration of the examination is long leading to high operational costs and (4) data fusion is difficult because of the different breathhold positions, scan geometries and non-isotropic spatial resolution. We hypothesize that image based respiratory self-navigation combined with image acceleration techniques will address the above challenges and allow improving the reliability and image quality of free-breathing (no breathholds) three-dimensional (3D) multi-contrast quantitative whole heart cardiac MRI. The proposed approach will enable non-invasive comprehensive cardiac examination with improved patient experience, higher diagnostic yield and improved cost effectiveness thereby improving the treatment and outcome of cardiovascular disease as outlined by the NHS white paper.

Planned Impact

Our goal is to maximise the impact of our work through dissemination of our ideas and results to the academic and clinical communities and potential industrial partners. The scientific methodology results from this research will be output as research publications in high-impact journals in the field of medical imaging and cardiology. Target journals will include Magnetic Resonance in Medicine, Circulation and Radiology. Dissemination will also take place through presentations at the major international academic conferences, especially the International Society for Magnetic Resonance in Medicine (ISMRM) and the Society for Cardiovascular Magnetic Resonance (SCMR).

A large amount of phantom and in-vivo cardiac MR data will be generated during the lifetime of this grant. We will protect any resulting intellectual property in consultation with KCL Enterprises. After the studies are published in scientific journals, and after they are patented if patenting is a viable option, this data will be made publicly available for research use. After completion of the project we also plan to share the developed software with other MR researchers and clinicians interested in our methods in order to validate the developed imaging methods at other medical centres and/or to further improve the methodology. Moreover, we will work closely with the manufactures to develop works in progress packages that can be easily shared with other universities or medical centres and that will also allow rapid translation of the developed methods into commercial products of the vendors. This will facilitate wider spread clinical use of this research and enhance the clinical impact of this research. The Division has a Master Research Agreement both with Philips and Siemens providing access to acquisition and reconstruction source code. Philips is currently funding, through our new EPSRC Centre for Doctoral Training (CDT) in Medical Imaging at KCL/Imperial College, one PhD student working on motion corrected 3D T1 mapping under the supervision of the PI, which is relevant to this project. Throughout the project we will post blogs about our latest results on the Division's blog (https://kingsimaging.wordpress.com/) to reach a larger community and to make our research results more accessible to a larger community. In addition, we will participate in the "Pint of Science" 3-day annual festival and present our research results in pubs around London to engage with local communities. We also plan to present to the Public and Patient Involvement Cardiovascular group at Guy's and St Thomas' Hospital to provide updates and receive patient feedback (http://www.guysandstthomasbrc.nihr.ac.uk/PatientsPublic/Getinvolved/Haveyoursay/Cardiovascular.aspx).

Publications

10 25 50

publication icon
Bratis K (2017) Clinical evaluation of three-dimensional late enhancement MRI. in Journal of magnetic resonance imaging : JMRI

publication icon
Bratis K (2017) 'Image-navigated 3-dimensional late gadolinium enhancement cardiovascular magnetic resonance imaging: feasibility and initial clinical results'. in Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance

publication icon
Correia T (2020) Accelerated high-resolution free-breathing 3D whole-heart T2-prepared black-blood and bright-blood cardiovascular magnetic resonance. in Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance

 
Description A new set of imaging tools for the non-invasive and radiation-free assessment of the anatomy of the coronary arteries and myocardium have been developed. These include the development of advanced respiratory motion correction methods which include beat-to-beat translational motion correction using image navigators (iNAV) and bin-to-bin non-rigid motion correction using the imaging data itself thereby allowing for shorter and predictable scan time resulting in improved image quality. In addition, novel imaging sequences were developed which allow the simultaneous visualisation of the coronary vessels, coronary thrombus, high intensity plaque and myocardial scar (BOOST sequence). These techniques now have been combined with advanced undersampling reconstruction techniques (ORCCA and PROST), which allow respiratory motion resolved reconstruction and highly undersampled reconstruction (3-4 fold) of non-rigid motion corrected whole heart coronary MR angiography (CMRA) datasets. To enable multi centric clinical validation we have developed works in progress packages (WIPs) together with Siemens Healthineers for our CMRA and BOOST sequence. The CMRA WIP also includes the option of performing free-breathing high-resolution motion corrected 3D myocardial viability imaging with and without black blood option, which is important for detection of small infarctions or arrhythmic substrate. The multi-contrast BOOST sequence has also been extended to allow joint T1/T2 mapping (Milotta G et al. MRM 2020) and was combined with the latest motion correction and image reconstruction developments to further increase image resolution and shorten scan time. We also have developed a motion corrected free running 3D whole heart T1 and joint T1/T2 mapping technique for simultaneous assessment of fibrosis and edema and which is based on a 3D radial trajectory and a low rank patched based reconstruction. All sequences are currently tested in patients with cardiovascular disease. Specifically, we have scanned 50 patients referred from the CTCA list with our high resolution CMRA protocol (0.9mm3) with the CMRA images approaching CT image quality but without the need for radiation or nephrotoxic contrast agents. The highlight of this clinical study is that all CMRA were completed successfully and that 97% of the proximal and 94% of the middle coronary segments were of diagnostic image quality (vs 99% and 98% for CTCA). Specificity and negative predicative value for identification of coronary artery disease were 93-98% and 95-100% for LM, LAD, RCA and LCx. We also have scanned 25 patients with congenital heart disease with our BOOST sequence which outperformed the clinical protocol in terms of image quality and diagnostic yield.
Exploitation Route Taken in combination, these developments will enable a better stratification of patients and therefore more personalized care of high-risk patients with more expensive technologies, as well as early diagnosis and better primary/secondary prevention.

Impacts
1. Significant progress towards clinically applicable and robust methods for anatomical and functional assessment of CAD.
2. Significant progress towards clinically applicable and robust methods for simultaneous assessment of coronary lumen integrity and high intensity coronary plaque.
3. Significant progress towards clinically applicable and robust methods for simultaneous assessment of coronary lumen and myocardial tissue integrity.
4. Significant progress towards clinically applicable and robust methods for high resolution myocardial viability imaging both with black blood and grey blood 3D LGE.
5. Significant progress towards clinically applicable and robust methods for high resolution myocardial T1 and T2 mapping.
6. Significant progress towards clinically applicable and robust methods for high resolution and accelerated coronary angiography using advanced motion correction and reconstruction methods.
Sectors Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology

 
Description The imaging sequences that we developed have been successfully used to diagnose CAD in over 200 patients at St Thomas' Hospital. In 2018, Siemens Healthineers decided to develop a pre-product software package (works-in-progress, or WIP) for coronary magnetic resonance angiography (CMRA), a necessary step prior to commercialisation. They incorporated this technology in their pre-product package across 10 medical centres worldwide (UK, US, Germany, France, China and Brazil) with a further 500+ patients diagnosed. Moreover, we extended the CMRA technology that we developed (iNAV, CASPR, motion corrected reconstruction with patch based low-rank denoising) to allow for detection of infarcted and viable myocardium using late gadolinium enhancement (LGE) MRI, which is now also included in the CMRA pre-product package further enhancing its clinical impact, as viability imaging is part of everyday routine cardiac MR exam, thus opening our technology to the entire cardiac MR market worldwide. The LGE option of our WIP is currently being used at the Charité in Berlin (>50 patients), at Erlangen University Hospital (>50 patients) and at Medneo MRI Centre in Berlin (>50 patients). Moreover, the WIP is also being used for the diagnosis of congenital heart disease in young children and adolescents in the Bambino Gesu Hospital in Rome (>50 patients). The excellent feedback from our clinical partners at St Thomas' and in Europe has led to multiple requests from several other sites in Europe, US and even in Latin America to use of technology for the diagnosis of CAD, myocardial infarction, atrial fibrillation and coronary vein visualization.
First Year Of Impact 2019
Sector Healthcare,Pharmaceuticals and Medical Biotechnology
 
Description Contrast-free Deep Myocardial Tissue Characterization with Cardiac MR Fingerprinting
Amount £932,527 (GBP)
Funding ID EP/V044087/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 08/2021 
End 09/2024
 
Description Detection of High-Risk Plaque with tropoelastin-specific and multicontrast coronary MRI
Amount £1,258,715 (GBP)
Funding ID RG/20/1/34802 
Organisation British Heart Foundation (BHF) 
Sector Charity/Non Profit
Country United Kingdom
Start 01/2021 
End 12/2025
 
Description Gadolinium-free multi-contrast 3D whole-heart MRI for improved management of patients with congenital heart disease
Amount £294,199 (GBP)
Organisation British Heart Foundation (BHF) 
Sector Charity/Non Profit
Country United Kingdom
Start  
 
Description Improved 3D Whole-heart MR Imaging for the Assessment of Cardiac Anatomy in Paediatric Congenital Heart Disease
Amount £57,869 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start  
 
Description Multi-parametric tissue characterisation of myocardial inflammation in autoimmune rheumatic diseases using cardiovascular magnetic resonance imaging
Amount £255,962 (GBP)
Funding ID FS/20/13/34857 
Organisation British Heart Foundation (BHF) 
Sector Charity/Non Profit
Country United Kingdom
Start  
 
Description Multidimensional and Multiparametric Quantitative Cardiac MRI from Continuous Free-Breathing Acquisition
Amount £565,581 (GBP)
Funding ID EP/P032311/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 08/2017 
End 08/2021
 
Description Simultaneous Non-Contrast Free Breathing 3D High Resolution Magnetic Resonance Coronary Artery Angiography and High-Risk Plaque Imaging
Amount £178,187 (GBP)
Funding ID FS/CRTF/20/24011 
Organisation British Heart Foundation (BHF) 
Sector Charity/Non Profit
Country United Kingdom
Start 09/2020 
End 09/2022
 
Description Charité Berlin 
Organisation Charité Campus Virchow - Hospital
Country Germany 
Sector Academic/University 
PI Contribution We have provided the partner with a cardiac PET-MR sequence and two multi-contrast whole heart MRI sequences for validation in patients with coronary artery disease.
Collaborator Contribution There are no contributions yet as we are waiting for the signing of the material transfer agreement.
Impact None so far.
Start Year 2018
 
Description Technical University Munich 
Organisation Technical University of Munich
Department Department of Nuclear Medicine
Country Germany 
Sector Academic/University 
PI Contribution We have developed a novel motion corrected cardiac PET-MR sequence that allows to simultaneously acquire a coronary MR angiogram and the PET list mode data whereby motion correction is performed in 2 steps; First translational motion correction is performed using the MR image navigator and secondly both the MR and PET data a assigned to 4-5 respiratory bins. From the binned MR data, non-rigid motion fields are estimated that are then used to correct both the MR and PET data on a bon-to-bin basis. We provided our collaborators in Munich with this novel sequence, wich has been published in MRM and they used this PET-MRI imaging sequence in patients with coronary occlusion.
Collaborator Contribution Our partners recruited 15 patients with coronary occlusion, as confirmed by x-ray angiography, and scanned those patients with our novel PET-MR sequence. They performed the PET image analysis before and after motion correction. This study demonstrated that MR based motion correction can improve PET myocardial viability images. In some patients a mural infarct was re-classified as non-transmural after motion correction. The study has been submitted to the European Journal of Nuclear Medicine.
Impact The results of this collaboration has been submitted to the European Journal of Nuclear Medicine.
Start Year 2017
 
Description University of Bordeaux 
Organisation University of Bordeaux
Country France 
Sector Academic/University 
PI Contribution We have provided our collaborators from the electrophysiology department with novel MRI pulse sequences (acquisition and reconstruction software) that we developed, which their are planning to use our MR sequences in large animal models and patients to perform myocardial tissue characterisation to guide interventional procedures. The collaborators are going to obtain our latest pre-product MR software to validate its usefulness for coronary angiography and late gadolinium enhancement MRI of the myocardium and atrial walls.
Collaborator Contribution So far our partner have not produced any published results.
Impact None so far.
Start Year 2018
 
Description University of Nantes 
Organisation University of Nantes
Country France 
Sector Academic/University 
PI Contribution We are providing our partner with novel MRI sequences that we have to developed in order to test those in patients with cardiovascular disease. The aim is to validate if our MRI sequences will allow detecting coronary intraplaque haemorrhage, which is considered a novel biomarker for future coronary events.
Collaborator Contribution No output yet, as we are waiting to sign a material transfer agreement.
Impact No outputs yet.
Start Year 2018
 
Title A method and apparatus for generating a T1-T2 map 
Description A method for generating a T1 and/or T2 map for a three-dimensional image volume comprises acquiring first, second, and third 3D images of the image volume using different MR acquisition parameters. Signal evolutions of the voxels through the first to third images are obtained by comparing voxel intensity levels of corresponding voxel locations in the three images. A simulation dictionary representing the signal evolutions for a number of different tissue parameter combinations is obtained by simulations e.g. using Bloch equations or extended phase graph (EPG) methods. The T1 and/or T2 map is generated by comparing the determined signal evolutions to entries in the dictionary and by finding, for each of the determined signal evolutions, the entry in the dictionary that best matches the determined signal evolution. The method may be used in cardiac imaging. The first image may use an inversion recovery preparation pulse, the second image may use no preparation pulse, and the third image may use a T2 preparation pulse. 
IP Reference GB2581168 
Protection Patent application published
Year Protection Granted 2020
Licensed No
Impact This methods enables high-resolution tissue characterization and has led to the creation of a Siemens co-funded PhD studentship.
 
Title METHOD AND APPARATUS FOR GENERATING A T1/T2 MAP 
Description A method and apparatus for generating a T1 or T2 map for a three-dimensional (3D) image volume of a subject. The method includes acquiring first, second, and third 3D images of the image volume of the subject. Signal evolutions of voxels through the first to third 3D images by comparing voxel intensity levels of corresponding voxel locations in the first, second, and third 3D images. A simulation dictionary representing the signal evolutions for a number of different tissue parameter combinations is obtained. The T1 or T2 map is generated by comparing the determined signal evolutions to entries in the dictionary and by finding, for each of the determined signal evolutions, the entry in the dictionary that best matches the determined signal evolution. 
IP Reference US2020249299 
Protection Patent application published
Year Protection Granted 2020
Licensed No
Impact This development had led to a joint patent application between King's College London and Siemens Healthineers. It also has led to the funding of a PhD project to extend the developed method to also quantify T2* and thus tissue iron concentration. Moreover, we are working towards a works in progress package together with Siemens Healthineers which is a prototype sequence required for commercialisation.
 
Title METHOD AND APPARATUS FOR RECONSTRUCTING MAGNETIC RESONANCE IMAGE DATA 
Description In a method for reconstructing magnetic resonance (MR) image data from k-space data, k-space data of an image region of a subject are provided to a computer that is also provided with multiple navigator signals for the image region of the subject. The computer sorts the k-space data into multiple bins, the multiple bins representing different motion states of the subject. For each of the multiple bins, the computer executes a compressed sensing procedure to reconstruct the MR image data from the k-space data in the respective bin. Execution of the compressed sensing procedure includes solving an optimization problem comprising a data consistency component and a transform sparsity component. Motion information is incorporated by the computer into at least one of the data consistency component and the transform sparsity component of the optimization problem. 
IP Reference US2019317172 
Protection Patent application published
Year Protection Granted 2019
Licensed No
Impact No licensing has occurred so far.
 
Title METHOD OF PERFORMING MAGNETIC RESONANCE IMAGING AND A MAGNETIC RESONANCE APPARATUS 
Description In a method of performing magnetic resonance (MR) imaging, an MR apparatus, and a computer-readable medium during a first cardiac cycle of a subject, a first imaging sequence is generated for application to a subject. The first imaging sequence has a preparatory pulse and an inversion recovery pulse following the preparatory pulse. First signals emitted from the subject in response to the first imaging sequence are detected, and first image data are generated based on the first signals. During a second cardiac cycle following the first cardiac cycle, a second imaging sequence is generated for application to the subject. The second imaging sequence has a preparatory pulse. Second signals emitted from the subject in response to the second imaging sequence are detected, and second image data are generated based on the second signals. 
IP Reference US2019064299 
Protection Patent application published
Year Protection Granted 2019
Licensed No
Impact No licensing has occurred so far.
 
Title METHOD OF RECONSTRUCTING MAGNETIC RESONANCE IMAGE DATA 
Description A method of reconstructing magnetic resonance (MR) image data from k-space data. The method includes obtaining k-space data of an image region of a subject; and reconstructing, using a sparse image coding procedure, the MR image data from the k-space data by performing an iterative optimization method. The optimization method includes a data consistency iteration step and a denoising iteration step applied to MR image data generated by the data consistency iteration step. The denoising iteration step incorporates a sparsifying operation to provide a sparse representation of the MR image data for the imaged region as an input to the data consistency iteration step. 
IP Reference US2019346522 
Protection Patent application published
Year Protection Granted 2019
Licensed No
Impact No licensing has occurred so far.
 
Title METHOD OF RECONSTRUCTING MAGNETIC RESONANCE IMAGE DATA 
Description A plurality of sets of k-space data each of the same image region of a subject but having different contrasts are obtained. A sparse image coding procedure is performed to reconstruct a plurality of MR images each corresponding to one of the sets of k-space data. This involves solving an optimization problem comprising a data consistency iteration step used to generate the reconstructed MR images; and a denoising iteration step applied to the reconstructed MR images generated during the data consistency iteration step. The denoising iteration step includes performing a 2D/3D block matching operation to identify similar patches across the reconstructed MR images, and using the similar patches across the reconstructed MR images in a sparsifying operation to provide sparse representations of the reconstructed MR images. The sparse representations are used as an input to the data consistency iteration step. 
IP Reference US2020241096 
Protection Patent application published
Year Protection Granted 2020
Licensed No
Impact This invention has enabled the development of many new multi-contrast cardiac MR sequences that were not feasible before by significantly speeding up image acquisition time by exploiting data redundancies in multiple dimensions.
 
Title Method and apparatus for correcting motion in K-Space data acquired in a magnetic resonance imaging procedure 
Description A method of correcting for motion in k-space data acquired during a magnetic resonance imaging (MRI) procedure comprising: acquiring a plurality of out-of-phase 2D image navigators 101 each representing, in the image domain, a 2D image of an image region of a subject; acquiring k-space data 102 representing, in the frequency domain, a plurality of 3D MRI images of the image region of the subject; and correcting for motion within the acquired k-space data 103 using motion information for the subject obtained from the plurality of out-of-phase 2D image navigators so as to generate motion corrected k-space 104; and an associated apparatus including a gradient system to apply a magnetic field gradient, an excitation system, a computing system in communication with the excitation system, and the gradient system for controlling these components. The k-space data may be sorted into bins (fig.2B, 213a,213b,213c) representing different motion states of the subject and a plurality of in-phase 2D image navigators may be acquired. 
IP Reference GB2582338 
Protection Patent application published
Year Protection Granted 2020
Licensed No
Impact This invention has enabled the detection of respiratory motion during cardiac MRI in multi-contrast images by providing fat images that are insensitive to contrast change in comparison to the typical water images.
 
Title Development of image navigator for respiratory motion correction in whole heart MRI 
Description We developed a new motion correction technique, which employs low resolution 2D images of the heart, prior to data acquisition to correct for respiratory motion on a heartbeat to heartbeat basis. The image navigator has been implemented both in the Philips and Siemens MR acquisition and reconstruction software and provides the clinician both with non-corrected and motion corrected images at the scanner console. 
Type Of Technology Software 
Year Produced 2018 
Impact The image navigator allows for 100% scan efficiency without any data rejection and makes whole heart MRI more reliable and scan time predictable compared to gating approaches, which can lead to prolonged scan times. The image navigator has been combined with several whole heart sequences (coronary MRA, late gadolinium enhancement, magnetisation transfer ratio, T1 and T2 mapping). We have now started several collaboration with academic institutions that are interested in evaluating this technology in patients with heart disease. 
 
Title Pre-product BOOST MRI sequence for simultaneous MR coronary angiography and plaque imaging 
Description We developed a pre-product with Siemens Helathineers which allows the simultaneous acquisition of a bright and black blood dataset that can be used for MR coronary angiography and plaque imaging. 
Type Of Technology New/Improved Technique/Technology 
Year Produced 2021 
Impact The BOOST sequence is currently validated against X-ray angiography and optical coherence tomography (OCT) for vulnerable plaque detection at Hammersmith Hospital. Correlation with X-ray and OCT are excellent suggesting that BOOST may be a non-invasive alternative. 
 
Title Pre-product MR sequence for coronary angiography and myocardial scar imaging 
Description We developed a novel MRI sequence both for coronary angiography and myocardial scar imaging that is currently tested worldwide by Siemens customers and is planned to become a product in 2022. 
Type Of Technology New/Improved Technique/Technology 
Year Produced 2020 
Impact This pre-product MR sequence has been requested by several leading physician scientists for clinical testing and is currently being evaluated at more than 5 centres worldwide (e.g. Charité Berlin, U Bordeaux, PUC in Beijing, U Bordeaux, U Nantes). 
 
Description Pint of Science activity in London 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact We presented our work at the Pint of Science in London where we had several talks about the use of medical imaging for the diagnosis and treatment of heart disease.
Year(s) Of Engagement Activity 2016
 
Description Royal Society Summer Science Festival 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact We took part of the Royal Society Summer Science Festival and had a stand demonstrating our research in cardiovascular imaging called "Heart in your Hands". Several hundred people attended our stand every day and all attendees were very engaged with our PhD students and postdocs who explained our research in lay words.
Year(s) Of Engagement Activity 2017