Imperial College London

ProfessorDavidFirmin

Faculty of MedicineNational Heart & Lung Institute

Professor of Biomedical Imaging
 
 
 
//

Contact

 

+44 (0)20 7351 8801d.firmin

 
 
//

Location

 

Cardiovascular MR UnitRoyal Brompton Campus

//

Summary

 

Publications

Publication Type
Year
to

401 results found

Yang G, Chen J, Gao Z, Li S, Ni H, Angelini E, Wong T, Mohiaddin R, Nyktari E, Wage R, Xu L, Zhang Y, Du X, Zhang H, Firmin D, Keegan Jet al., Simultaneous Left Atrium Anatomy and Scar Segmentations via Deep Learning in Multiview Information with Attention, Future Generation Computer Systems: the international journal of grid computing: theory, methods and applications, ISSN: 0167-739X

Journal article

Li L, Wu F, Yang G, Xu L, Wong T, Mohiaddin R, Firmin D, Keegan J, Zhuang Xet al., 2020, Atrial scar quantification via multi-scale CNN in the graph-cuts framework, Medical Image Analysis, Vol: 60, ISSN: 1361-8415

Late gadolinium enhancement magnetic resonance imaging (LGE MRI) appears to be a promising alternative for scarassessment in patients with atrial fibrillation (AF). Automating the quantification and analysis of atrial scars can bechallenging due to the low image quality. In this work, we propose a fully automated method based on the graph-cutsframework, where the potentials of the graph are learned on a surface mesh of the left atrium (LA) using a multi-scaleconvolutional neural network (MS-CNN). For validation, we have included fifty-eight images with manual delineations.MS-CNN, which can efficiently incorporate both the local and global texture information of the images, has been shownto evidently improve the segmentation accuracy of the proposed graph-cuts based method. The segmentation could befurther improved when the contribution between the t-link and n-link weights of the graph is balanced. The proposedmethod achieves a mean accuracy of 0.856 ± 0.033 and mean Dice score of 0.702 ± 0.071 for LA scar quantification.Compared to the conventional methods, which are based on the manual delineation of LA for initialization, our methodis fully automatic and has demonstrated significantly better Dice score and accuracy (p < 0.01). The method is promisingand can be potentially useful in diagnosis and prognosis of AF.

Journal article

Zhuang X, Li L, Payer C, Stern D, Urschler M, Heinrich M, Oster J, Wang C, Smedby O, Bian C, Yang X, Heng P-A, Mortazi A, Bagci U, Yang G, Sun C, Galisot G, Ramel J-Y, Brouard T, Tong Q, Si W, Liao X, Zeng G, Shi Z, Zheng G, Wang C, MacGillivray T, Newby D, Rhode K, Ourselin S, Mohiaddin R, Keegan J, Firmin D, Yang Get al., 2019, Evaluation of algorithms for multi-modality whole heart segmentation: An open-access grand challenge, Medical Image Analysis, Vol: 58, ISSN: 1361-8415

Knowledge of whole heart anatomy is a prerequisite for many clinical applications. Whole heart segmentation (WHS),which delineates substructures of the heart, can be very valuable for modeling and analysis of the anatomy and functionsof the heart. However, automating this segmentation can be challenging due to the large variation of the heart shape,and different image qualities of the clinical data. To achieve this goal, an initial set of training data is generally neededfor constructing priors or for training. Furthermore, it is difficult to perform comparisons between different methods,largely due to differences in the datasets and evaluation metrics used. This manuscript presents the methodologiesand evaluation results for the WHS algorithms selected from the submissions to the Multi-Modality Whole Heart Segmentation (MM-WHS) challenge, in conjunction with MICCAI 2017. The challenge provided 120 three-dimensionalcardiac images covering the whole heart, including 60 CT and 60 MRI volumes, all acquired in clinical environmentswith manual delineation. Ten algorithms for CT data and eleven algorithms for MRI data, submitted from twelvegroups, have been evaluated. The results showed that the performance of CT WHS was generally better than thatof MRI WHS. The segmentation of the substructures for different categories of patients could present different levelsof challenge due to the difference in imaging and variations of heart shapes. The deep learning (DL)-based methodsdemonstrated great potential, though several of them reported poor results in the blinded evaluation. Their performance could vary greatly across different network structures and training strategies. The conventional algorithms,mainly based on multi-atlas segmentation, demonstrated good performance, though the accuracy and computationalefficiency could be limited. The challenge, including provision of the annotated training data and the blinded evaluation for submitted algorithms on the test data, conti

Journal article

Khalique Z, Scott AD, Ferreira PF, Nielles-Vallespin S, Firmin DN, Pennell DJet al., 2019, Diffusion tensor cardiovascular magnetic resonance in hypertrophic cardiomyopathy: a comparison of motion-compensated spin echo and stimulated echo techniques, Magnetic Resonance Materials in Physics, Biology and Medicine, ISSN: 0968-5243

ObjectivesDiffusion tensor cardiovascular magnetic resonance (DT-CMR) interrogates myocardial microstructure. Two frequently used in vivo DT-CMR techniques are motion-compensated spin echo (M2-SE) and stimulated echo acquisition mode (STEAM). Whilst M2-SE is strain-insensitive and signal to noise ratio efficient, STEAM has a longer diffusion time and motion compensation is unnecessary. Here we compare STEAM and M2-SE DT-CMR in patients.Materials and methodsBiphasic DT-CMR using STEAM and M2-SE, late gadolinium imaging and pre/post gadolinium T1-mapping were performed in a mid-ventricular short-axis slice, in ten hypertrophic cardiomyopathy (HCM) patients at 3 T.ResultsAdequate quality data were obtained from all STEAM, but only 7/10 (systole) and 4/10 (diastole) M2-SE acquisitions. Compared with STEAM, M2-SE yielded higher systolic mean diffusivity (MD) (p = 0.02) and lower fractional anisotropy (FA) (p = 0.02, systole). Compared with segments with neither hypertrophy nor late gadolinium, segments with both had lower systolic FA using M2-SE (p = 0.02) and trend toward higher MD (p = 0.1). The negative correlation between FA and extracellular volume fraction was stronger with STEAM than M2-SE (r2 = 0.29, p < 0.001 STEAM vs. r2 = 0.10, p = 0.003 M2-SE).DiscussionIn HCM, only STEAM reliably assesses biphasic myocardial microstructure. Higher MD and lower FA from M2-SE reflect the shorter diffusion times. Further work will relate DT-CMR parameters and microstructural changes in disease.

Journal article

Chen J, Zhang H, Zhang Y, Zhao S, Mohiaddin R, Wong T, Firmin D, Yang G, Keegan Jet al., 2019, Discriminative consistent domain generation for semi-supervised learning, International Conference on Medical Image Computing and Computer-Assisted Intervention, Publisher: Springer International Publishing, Pages: 595-604, ISSN: 0302-9743

Deep learning based task systems normally rely on a large amount of manually labeled training data, which is expensive to obtain and subject to operator variations. Moreover, it does not always hold that the manually labeled data and the unlabeled data are sitting in the same distribution. In this paper, we alleviate these problems by proposing a discriminative consistent domain generation (DCDG) approach to achieve a semi-supervised learning. The discriminative consistent domain is achieved by a double-sided domain adaptation. The double-sided domain adaptation aims to make a fusion of the feature spaces of labeled data and unlabeled data. In this way, we can fit the differences of various distributions between labeled data and unlabeled data. In order to keep the discriminativeness of generated consistent domain for the task learning, we apply an indirect learning for the double-sided domain adaptation. Based on the generated discriminative consistent domain, we can use the unlabeled data to learn the task model along with the labeled data via a consistent image generation. We demonstrate the performance of our proposed DCDG on the late gadolinium enhancement cardiac MRI (LGE-CMRI) images acquired from patients with atrial fibrillation in two clinical centers for the segmentation of the left atrium anatomy (LA) and proximal pulmonary veins (PVs). The experiments show that our semi-supervised approach achieves compelling segmentation results, which can prove the robustness of DCDG for the semi-supervised learning using the unlabeled data along with labeled data acquired from a single center or multicenter studies.

Conference paper

Nielles-Vallespin S, Scott A, Ferreira P, Khalique Z, Pennell D, Firmin Det al., 2019, Cardiac Diffusion: Technique and Practical Applications., J Magn Reson Imaging

The 3D microarchitecture of the cardiac muscle underlies the mechanical and electrical properties of the heart. Cardiomyocytes are arranged helically through the depth of the wall, and their shortening leads to macroscopic torsion, twist, and shortening during cardiac contraction. Furthermore, cardiomyocytes are organized in sheetlets separated by shear layers, which reorientate, slip, and shear during macroscopic left ventricle (LV) wall thickening. Cardiac diffusion provides a means for noninvasive interrogation of the 3D microarchitecture of the myocardium. The fundamental principle of MR diffusion is that an MRI signal is attenuated by the self-diffusion of water in the presence of large diffusion-encoding gradients. Since water molecules are constrained by the boundaries in biological tissue (cell membranes, collagen layers, etc.), depicting their diffusion behavior elucidates the shape of the myocardial microarchitecture they are embedded in. Cardiac diffusion therefore provides a noninvasive means to understand not only the dynamic changes in cardiac microstructure of healthy myocardium during cardiac contraction but also the pathophysiological changes in the presence of disease. This unique and innovative technology offers tremendous potential to enable improved clinical diagnosis through novel microstructural and functional assessment. in vivo cardiac diffusion methods are immediately translatable to patients, opening new avenues for diagnostic investigation and treatment evaluation in a range of clinically important cardiac pathologies. This review article describes the 3D microstructure of the LV, explains in vivo and ex vivo cardiac MR diffusion acquisition and postprocessing techniques, as well as clinical applications to date. Level of Evidence: 1 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2019.

Journal article

Gulati A, Ismail TF, Ali A, Hsu L-Y, Goncalves C, Ismail NA, Krishnathasan K, Davendralingam N, Ferreira P, Halliday BP, Jones DA, Wage R, Newsome S, Gatehouse P, Firmin D, Jabbour A, Assomull RG, Mathur A, Pennell DJ, Arai AE, Prasad SKet al., 2019, Microvascular Dysfunction in Dilated Cardiomyopathy A Quantitative Stress Perfusion Cardiovascular Magnetic Resonance Study, JACC-CARDIOVASCULAR IMAGING, Vol: 12, Pages: 1699-1708, ISSN: 1936-878X

Journal article

Khalique Z, Ferreira P, Scott A, Nielles-Vallespin S, Firmin D, Pennell Det al., Diffusion tensor cardiovascular magnetic resonance: a clinical perspective, JACC: Cardiovascular Imaging, ISSN: 1936-878X

Imaging the heart is central to cardiac phenotyping but in clinical practice this has been restricted to macroscopic interrogation. Diffusion tensor cardiovascular magnetic resonance (DT-CMR) is a novel, non-invasive technique which is beginning to unlock details of this microstructure in humans in-vivo. DT-CMR demonstrates the helical cardiomyocyte arrangement that drives rotation and torsion. Sheetlets (functional units of cardiomyocytes, separated by shear layers) have been shown to reorientate between diastole and systole, revealing how microstructural function facilitates cardiac thickening. Measures of tissue diffusion can also be made; fractional anisotropy (a measure of myocyte organisation) and mean diffusivity (a measure of myocyte packing). Abnormal myocyte orientation and sheetlet function has been demonstrated in congenital heart disease, cardiomyopathy and after myocardial infarction. It is too early to predict the clinical importance of DT-CMR, but such unique in-vivo information will likely prove valuable in early diagnosis and risk prediction of cardiac dysfunction and arrhythmias.

Journal article

Zhang N, Yang G, Gao Z, Xu C, Zhang Y, Shi R, Keegan J, Xu L, Zhang H, Fan Z, Firmin Det al., 2019, Deep learning for diagnosis of chronic myocardial infarction on nonenhanced cardiac cine MRI, Radiology, Vol: 294, Pages: 52-60, ISSN: 0033-8419

BackgroundRenal impairment is common in patients with coronary artery disease and, if severe, late gadolinium enhancement (LGE) imaging for myocardial infarction (MI) evaluation cannot be performed.PurposeTo develop a fully automatic framework for chronic MI delineation via deep learning on non–contrast material–enhanced cardiac cine MRI.Materials and MethodsIn this retrospective single-center study, a deep learning model was developed to extract motion features from the left ventricle and delineate MI regions on nonenhanced cardiac cine MRI collected between October 2015 and March 2017. Patients with chronic MI, as well as healthy control patients, had both nonenhanced cardiac cine (25 phases per cardiac cycle) and LGE MRI examinations. Eighty percent of MRI examinations were used for the training data set and 20% for the independent testing data set. Chronic MI regions on LGE MRI were defined as ground truth. Diagnostic performance was assessed by analysis of the area under the receiver operating characteristic curve (AUC). MI area and MI area percentage from nonenhanced cardiac cine and LGE MRI were compared by using the Pearson correlation, paired t test, and Bland-Altman analysis.ResultsStudy participants included 212 patients with chronic MI (men, 171; age, 57.2 years ± 12.5) and 87 healthy control patients (men, 42; age, 43.3 years ± 15.5). Using the full cardiac cine MRI, the per-segment sensitivity and specificity for detecting chronic MI in the independent test set was 89.8% and 99.1%, respectively, with an AUC of 0.94. There were no differences between nonenhanced cardiac cine and LGE MRI analyses in number of MI segments (114 vs 127, respectively; P = .38), per-patient MI area (6.2 cm2 ± 2.8 vs 5.5 cm2 ± 2.3, respectively; P = .27; correlation coefficient, r = 0.88), and MI area percentage (21.5% ± 17.3 vs 18.5% ± 15.4; P = .17; correlation coefficient, r = 0.89).ConclusionThe proposed deep learning f

Journal article

Tayal U, Wage R, Ferreira P, Nielles-Vallespin S, Epstein F, Auger D, Zhong X, Pennell D, Firmin D, Scott A, Prasad Set al., 2019, The feasibility of a novel limited field of view spiral cine DENSE sequence to assess myocardial strain in dilated cardiomyopathy, Magnetic Resonance Materials in Physics, Biology and Medicine, Vol: 32, Pages: 317-329, ISSN: 0968-5243

ObjectiveDevelop an accelerated cine displacement encoding with stimulated echoes (DENSE) cardiovascular magnetic resonance (CMR) sequence to enable clinically feasible myocardial strain evaluation in patients with dilated cardiomyopathy (DCM).Materials and methodsA spiral cine DENSE sequence was modified by limiting the field of view in two dimensions using in-plane slice-selective pulses in the stimulated echo. This reduced breath hold duration from 20RR to 14RR intervals. Following phantom and pilot studies, the feasibility of the sequence to assess peak radial, circumferential, and longitudinal strain was tested in control subjects (n = 18) and then applied in DCM patients (n = 29).ResultsDENSE acquisition was possible in all participants. Elements of the data were not analysable in 1 control (6%) and 4 DCM r(14%) subjects due to off-resonance or susceptibility artefacts and low signal-to-noise ratio. Peak radial, circumferential, short-axis contour strain and longitudinal strain was reduced in DCM patients (p < 0.001 vs. controls) and strain measurements correlated with left ventricular ejection fraction (with circumferential strain r = − 0.79, p < 0.0001; with vertical long-axis strain r = − 0.76, p < 0.0001). All strain measurements had good inter-observer agreement (ICC > 0.80), except peak radial strain.DiscussionWe demonstrate the feasibility of CMR strain assessment in healthy controls and DCM patients using an accelerated cine DENSE technique. This may facilitate integration of strain assessment into routine CMR studies.

Journal article

Stoeck CT, Scott AD, Ferreira PF, Tunnicliffe EM, Teh I, Nielles-Vallespin S, Moulin K, Sosnovik DE, Viallon M, Croisille P, Kozerke S, Firmin DN, Ennis DB, Schneider JEet al., 2019, Motion-induced signal loss in In vivo cardiac diffusion-weighted imaging, Journal of Magnetic Resonance Imaging, ISSN: 1053-1807

Journal article

Rose JN, Nielles-Vallespin S, Ferreira PF, Firmin DN, Scott AD, Doorly DJet al., 2019, Novel insights into in-vivo diffusion tensor cardiovascular magnetic resonance using computational modelling and a histology-based virtual microstructure, Magnetic Resonance in Medicine, Vol: 81, Pages: 2759-2773, ISSN: 0740-3194

PurposeTo develop histology‐informed simulations of diffusion tensor cardiovascular magnetic resonance (DT‐CMR) for typical in‐vivo pulse sequences and determine their sensitivity to changes in extra‐cellular space (ECS) and other microstructural parameters.MethodsWe synthesised the DT‐CMR signal from Monte Carlo random walk simulations. The virtual tissue was based on porcine histology. The cells were thickened and then shrunk to modify ECS. We also created idealised geometries using cuboids in regular arrangement, matching the extra‐cellular volume fraction (ECV) of 16–40%. The simulated voxel size was 2.8 × 2.8 × 8.0 mm3 for pulse sequences covering short and long diffusion times: Stejskal–Tanner pulsed‐gradient spin echo, second‐order motion‐compensated spin echo, and stimulated echo acquisition mode (STEAM), with clinically available gradient strengths.ResultsThe primary diffusion tensor eigenvalue increases linearly with ECV at a similar rate for all simulated geometries. Mean diffusivity (MD) varies linearly, too, but is higher for the substrates with more uniformly distributed ECS. Fractional anisotropy (FA) for the histology‐based geometry is higher than the idealised geometry with low sensitivity to ECV, except for the long mixing time of the STEAM sequence. Varying the intra‐cellular diffusivity (DIC) results in large changes of MD and FA. Varying extra‐cellular diffusivity or using stronger gradients has minor effects on FA. Uncertainties of the primary eigenvector orientation are reduced using STEAM.ConclusionsWe found that the distribution of ECS has a measurable impact on DT‐CMR parameters. The observed sensitivity of MD and FA to ECV and DIC has potentially interesting applications for interpreting in‐vivo DT‐CMR parameters.

Journal article

Khalique Z, Ferreira PF, Scott AD, Nielles-Vallespin S, Wage R, Martinez-Naharro A, Fontana M, Hawkins PN, Firmin DN, Pennell DJet al., 2019, DIFFUSION TENSOR CARDIOVASCULAR MAGNETIC RESONANCE IN CARDIAC AMYLOIDOSIS, Annual Meeting of the British-Society-of-Cardiovascular-Magnetic-Resonance (BSCMR), Publisher: BMJ PUBLISHING GROUP, Pages: A6-A7, ISSN: 1355-6037

Conference paper

Gorodezky M, Ferreira P, Nielles-Vallespin S, Gatehouse P, Pennell D, Scott A, Firmin Det al., 2019, High resolution in-vivo DT-CMR using an interleaved variable density spiral STEAM sequence, Magnetic Resonance in Medicine, Vol: 81, Pages: 1580-1594, ISSN: 0740-3194

Purpose: Diffusion tensor cardiovascular magnetic resonance (DT-CMR) has a limited spatial resolution. Thepurpose of this study was to demonstrate high-resolution DT-CMR using a segmented variable density spiralsequence with correction for motion, off-resonance and T2* related blurring.Methods: A single-shot STEAM EPI DT-CMR sequence at 2.8x2.8x8mm3 and 1.8x1.8x8mm3 was compared to asingle shot spiral at 2.8x2.8x8mm3 and an interleaved spiral sequence at 1.8x1.8x8mm3resolution in 10 healthyvolunteers at peak-systole and diastasis. Motion-induced phase was corrected using the densely sampledcentral k-space data of the spirals. STEAM field maps and T2* measures were obtained using a pair ofstimulated echoes each with a double spiral readout, the first used to correct the motion-induced phase of thesecond.Results: The high resolution spiral sequence produced similar DT-CMR results and quality measures to thestandard resolution sequence in both cardiac phases. Residual differences in fractional anisotropy and helixangle gradient between the resolutions could be due to spatial resolution and/or signal to noise ratio. The dataquality increased after both motion-induced phase correction and off-resonance correction and sharpnessincreased after T2* correction. The high resolution EPI sequence failed to provide sufficient data quality forDT-CMR reconstruction.Conclusion: In this study an in-vivo DT-CMR acquisition at 1.8x1.8mm2in-plane resolution was demonstratedusing a segmented spiral STEAM sequence. The motion-induced phase and off-resonance corrections areessential for high resolution spiral DT-CMR. Segmented variable density spiral STEAM was found to be theoptimal method for acquiring high resolution DT-CMR data.

Journal article

Li L, Yang G, Wu F, Wong T, Mohiaddin R, Firmin D, Keegan J, Xu L, Zhuang Xet al., 2019, Atrial Scar Segmentation via Potential Learning in the Graph-Cut Framework, Pages: 152-160, ISSN: 0302-9743

© 2019, Springer Nature Switzerland AG. Late Gadolinium Enhancement Magnetic Resonance Imaging (LGE MRI) emerges as a routine scan for patients with atrial fibrillation (AF). However, due to the low image quality automating the quantification and analysis of the atrial scars is challenging. In this study, we proposed a fully automated method based on the graph-cut framework, where the potential of the graph is learned on a surface mesh of the left atrium (LA), using an equidistant projection and a deep neural network (DNN). For validation, we employed 100 datasets with manual delineation. The results showed that the performance of the proposed method was improved and converged with respect to the increased size of training patches, which provide important features of the structural and texture information learned by the DNN. The segmentation could be further improved when the contribution from the t-link and n-link is balanced, thanks to the inter-relationship learned by the DNN for the graph-cut algorithm. Compared with the existing methods which mostly acquired an initialization from manual delineation of the LA or LA wall, our method is fully automated and has demonstrated great potentials in tackling this task. The accuracy of quantifying the LA scars using the proposed method was 0.822, and the Dice score was 0.566. The results are promising and the method can be useful in diagnosis and prognosis of AF.

Conference paper

Yang G, Chen J, Gao Z, Zhang H, Ni H, Angelini E, Mohiaddin R, Wong T, Keegan J, Firmin Det al., 2018, Multiview sequential learning and dilated residual learning for a fully automatic delineation of the left atrium and pulmonary veins from late gadolinium-enhanced cardiac MRI images, 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Publisher: IEEE, Pages: 1123-1127, ISSN: 1557-170X

Accurate delineation of heart substructures is a prerequisite for abnormality detection, for making quantitative and functional measurements, and for computer-aided diagnosis and treatment planning. Late Gadolinium-Enhanced Cardiac MRI (LGE-CMRI) is an emerging imaging technology for myocardial infarction or scar detection based on the differences in the volume of residual gadolinium distribution between scar and healthy tissues. While LGE-CMRI is a well-established non-invasive tool for detecting myocardial scar tissues in the ventricles, its application to left atrium (LA) imaging is more challenging due to its very thin wall of the LA and poor quality images, which may be produced because of motion artefacts and low signal-to-noise ratio. As the LGE-CMRI scan is designed to highlight scar tissues by altering the gadolinium kinetics, the anatomy among different heart substructures has less distinguishable boundaries. An accurate, robust and reproducible method for LA segmentation is highly in demand because it can not only provide valuable information of the heart function but also be helpful for the further delineation of scar tissue and measuring the scar percentage. In this study, we proposed a novel deep learning framework working on LGE-CMRI images directly by combining sequential learning and dilated residual learning to delineate LA and pulmonary veins fully automatically. The achieved results showed accurate segmentation results compared to the state-of-the-art methods. The proposed framework leads to an automatic generation of a patient-specific model that can potentially enable an objective atrial scarring assessment for the atrial fibrillation patients.

Conference paper

Khalique Z, Ferreira P, Scott A, Nielles-Vallespin S, Wage R, Firmin D, Pennell Det al., 2018, Diffusion Tensor Cardiovascular Magnetic Resonance of Microstructural Recovery in Dilated Cardiomyopathy, JACC: Cardiovascular Imaging, Vol: 11, Pages: 1548-1550, ISSN: 1936-878X

Journal article

Schlemper J, Yang G, Ferreira P, Scott A, McGill LA, Khalique Z, Gorodezky M, Roehl M, Keegan J, Pennell D, Firmin D, Rueckert Det al., 2018, Stochastic deep compressive sensing for the reconstruction of diffusion tensor cardiac MRI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol: 11070 LNCS, Pages: 295-303, ISSN: 0302-9743

© Springer Nature Switzerland AG 2018. Understanding the structure of the heart at the microscopic scale of cardiomyocytes and their aggregates provides new insights into the mechanisms of heart disease and enables the investigation of effective therapeutics. Diffusion Tensor Cardiac Magnetic Resonance (DT-CMR) is a unique non-invasive technique that can resolve the microscopic structure, organisation, and integrity of the myocardium without the need for exogenous contrast agents. However, this technique suffers from relatively low signal-to-noise ratio (SNR) and frequent signal loss due to respiratory and cardiac motion. Current DT-CMR techniques rely on acquiring and averaging multiple signal acquisitions to improve the SNR. Moreover, in order to mitigate the influence of respiratory movement, patients are required to perform many breath holds which results in prolonged acquisition durations (e.g., ~ 30 min using the existing technology). In this study, we propose a novel cascaded Convolutional Neural Networks (CNN) based compressive sensing (CS) technique and explore its applicability to improve DT-CMR acquisitions. Our simulation based studies have achieved high reconstruction fidelity and good agreement between DT-CMR parameters obtained with the proposed reconstruction and fully sampled ground truth. When compared to other state-of-the-art methods, our proposed deep cascaded CNN method and its stochastic variation demonstrated significant improvements. To the best of our knowledge, this is the first study using deep CNN based CS for the DT-CMR reconstruction. In addition, with relatively straightforward modifications to the acquisition scheme, our method can easily be translated into a method for online, at-the-scanner reconstruction enabling the deployment of accelerated DT-CMR in various clinical applications.

Journal article

Chen J, Yang G, Gao Z, Ni H, Angelini E, Mohiaddin R, Wong T, Zhang Y, Du X, Zhang H, Keegan J, Firmin Det al., 2018, Multiview two-task recursive attention model for left atrium and atrial scars segmentation, Medical Image Computing and Computer Assisted Intervention – MICCAI 2018, Publisher: Springer, Pages: 455-463, ISSN: 0302-9743

Late Gadolinium Enhanced Cardiac MRI (LGE-CMRI) for detecting atrial scars in atrial fibrillation (AF) patients has recently emerged as a promising technique to stratify patients, guide ablation therapy and predict treatment success. Visualisation and quantification of scar tissues require a segmentation of both the left atrium (LA) and the high intensity scar regions from LGE-CMRI images. These two segmentation tasks are challenging due to the cancelling of healthy tissue signal, low signal-to-noise ratio and often limited image quality in these patients. Most approaches require manual supervision and/or a second bright-blood MRI acquisition for anatomical segmentation. Segmenting both the LA anatomy and the scar tissues automatically from a single LGE-CMRI acquisition is highly in demand. In this study, we proposed a novel fully automated multiview two-task (MVTT) recursive attention model working directly on LGE-CMRI images that combines a sequential learning and a dilated residual learning to segment the LA (including attached pulmonary veins) and delineate the atrial scars simultaneously via an innovative attention model. Compared to other state-of-the-art methods, the proposed MVTT achieves compelling improvement, enabling to generate a patient-specific anatomical and atrial scar assessment model.

Conference paper

Wu F, Li L, Yang G, Wong T, Mohiaddin R, Firmin D, Keegan J, Xu L, Zhuang Xet al., 2018, Atrial Fibrosis Quantification Based on Maximum Likelihood Estimator of Multivariate Images, 21st International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI2018), Pages: 604-612, ISSN: 0302-9743

© 2018, Springer Nature Switzerland AG. We present a fully-automated segmentation and quantification of the left atrial (LA) fibrosis and scars combining two cardiac MRIs, one is the target late gadolinium-enhanced (LGE) image, and the other is an anatomical MRI from the same acquisition session. We formulate the joint distribution of images using a multivariate mixture model (MvMM), and employ the maximum likelihood estimator (MLE) for texture classification of the images simultaneously. The MvMM can also embed transformations assigned to the images to correct the misregistration. The iterated conditional mode algorithm is adopted for optimization. This method first extracts the anatomical shape of the LA, and then estimates a prior probability map. It projects the resulting segmentation onto the LA surface, for quantification and analysis of scarring. We applied the proposed method to 36 clinical data sets and obtained promising results (Accuracy: 0.809±150, Dice: 0.556±187). We compared the method with the conventional algorithms and showed an evidently and statistically better performance (p < 0.03).

Conference paper

Seitzer M, Yang G, Schlemper J, Oktay O, Würfl T, Christlein V, Wong T, Mohiaddin R, Firmin D, Keegan J, Rueckert D, Maier Aet al., 2018, Adversarial and perceptual refinement for compressed sensing MRI reconstruction, 21st International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI2018), Pages: 232-240, ISSN: 0302-9743

© Springer Nature Switzerland AG 2018. Deep learning approaches have shown promising performance for compressed sensing-based Magnetic Resonance Imaging. While deep neural networks trained with mean squared error (MSE) loss functions can achieve high peak signal to noise ratio, the reconstructed images are often blurry and lack sharp details, especially for higher undersampling rates. Recently, adversarial and perceptual loss functions have been shown to achieve more visually appealing results. However, it remains an open question how to (1) optimally combine these loss functions with the MSE loss function and (2) evaluate such a perceptual enhancement. In this work, we propose a hybrid method, in which a visual refinement component is learnt on top of an MSE loss-based reconstruction network. In addition, we introduce a semantic interpretability score, measuring the visibility of the region of interest in both ground truth and reconstructed images, which allows us to objectively quantify the usefulness of the image quality for image post-processing and analysis. Applied on a large cardiac MRI dataset simulated with 8-fold undersampling, we demonstrate significant improvements (p<0.01) over the state-of-the-art in both a human observer study and the semantic interpretability score.

Conference paper

Pennell DJ, Khalique Z, Ferreira PF, Scott AD, Nielles-Vallespin S, Kilner PJ, Kutys R, Romero M, Arai AE, Firmin DNet al., 2018, Deranged myocyte microstructure in situs inversus totalis demonstrated by diffusion tensor cardiovascular magnetic resonance, JACC: Cardiovascular Imaging, Vol: 11, Pages: 1360-1362, ISSN: 1936-878X

Journal article

Gorodezky M, Scott AD, Ferreira PF, Nielles-Vallespin S, Pennell DJ, Firmin DNet al., 2018, Diffusion tensor cardiovascular magnetic resonance with a spiral trajectory: An in vivo comparison of echo planar and spiral stimulated echo sequences, Magnetic Resonance in Medicine, Vol: 80, Pages: 648-654, ISSN: 0740-3194

PURPOSE: Diffusion tensor cardiovascular MR (DT-CMR) using stimulated echo acquisition mode (STEAM) with echo-planar-imaging (EPI) readouts is a low signal-to-noise-ratio (SNR) technique and therefore typically has a low spatial resolution. Spiral trajectories are more efficient than EPI, and could increase the SNR. The purpose of this study was to compare the performance of a novel STEAM spiral DT-CMR sequence with an equivalent established EPI technique. METHODS: A STEAM DT-CMR sequence was implemented with a spiral readout and a reduced field of view. An in vivo comparison of DT-CMR parameters and data quality between EPI and spiral was performed in 11 healthy volunteers imaged in peak systole and diastasis at 3 T. The SNR was compared in a phantom and in vivo. RESULTS: There was a greater than 49% increase in the SNR in vivo and in the phantom measurements (in vivo septum, systole: SNREPI  = 8.0 ± 2.2, SNRspiral  = 12.0 ± 2.7; diastasis: SNREPI  = 8.1 ± 1.6, SNRspiral  = 12.0 ± 3.7). There were no significant differences in helix angle gradient (HAG) (systole: HAGEPI  = -0.79 ± 0.07 °/%; HAGspiral  = -0.74 ± 0.16 °/%; P = 0.11; diastasis: HAGEPI  = -0.63 ± 0.05 °/%; HAGspiral  = -0.56 ± 0.14 °/%; P = 0.20), mean diffusivity (MD) in systole (MDEPI  = 0.99 ± 0.06 × 10-3 mm2 /s, MDspiral  = 1.00 ± 0.09 × 10-3 mm2 /s, P = 0.23) and secondary eigenvector angulation (E2A) (systole: E2AEPI  = 61 ± 10 °; E2Aspiral  = 63 ± 10 °; P&thi

Journal article

Yang G, Yu S, Hao D, Slabaugh G, Dragotti PL, Ye X, Liu F, Arridge S, Keegan J, Guo Y, Firmin Det al., 2018, DAGAN: deep de-aliasing generative adversarial networks for fast compressed sensing MRI reconstruction, IEEE Transactions on Medical Imaging, Vol: 37, Pages: 1310-1321, ISSN: 0278-0062

Compressed Sensing Magnetic Resonance Imaging (CS-MRI) enables fast acquisition, which is highly desirable for numerous clinical applications. This can not only reduce the scanning cost and ease patient burden, but also potentially reduce motion artefacts and the effect of contrast washout, thus yielding better image quality. Different from parallel imaging based fast MRI, which utilises multiple coils to simultaneously receive MR signals, CS-MRI breaks the Nyquist-Shannon sampling barrier to reconstruct MRI images with much less required raw data. This paper provides a deep learning based strategy for reconstruction of CS-MRI, and bridges a substantial gap between conventional non-learning methods working only on data from a single image, and prior knowledge from large training datasets. In particular, a novel conditional Generative Adversarial Networks-based model (DAGAN) is proposed to reconstruct CS-MRI. In our DAGAN architecture, we have designed a refinement learning method to stabilise our U-Net based generator, which provides an endto-end network to reduce aliasing artefacts. To better preserve texture and edges in the reconstruction, we have coupled the adversarial loss with an innovative content loss. In addition, we incorporate frequency domain information to enforce similarity in both the image and frequency domains. We have performed comprehensive comparison studies with both conventional CSMRI reconstruction methods and newly investigated deep learning approaches. Compared to these methods, our DAGAN method provides superior reconstruction with preserved perceptual image details. Furthermore, each image is reconstructed in about 5 ms, which is suitable for real-time processing.

Journal article

Khalique Z, Ferreira PF, Scott AD, Nielles-Vallespin S, Wage R, Firmin DN, Pennell DJet al., 2018, ASSESSMENT OF THE MICROSTRUCTURE IN RECOVERED DILATED CARDIOMYOPATHY WITH DIFFUSION TENSOR CARDIOVASCULAR MAGNETIC RESONANCE, Joint Meeting of the British-Society-of-Cardiovascular-Imaging/British-Society-of-Cardiovascular-CT, British-Society-of-Cardiovascular-Magnetic-Resonance and British-Nuclear-Cardiac-Society on British Cardiovascular Imaging, Publisher: BMJ PUBLISHING GROUP, Pages: A6-A7, ISSN: 1355-6037

Conference paper

Yang G, Zhuang X, Khan H, Haldar S, Nyktari E, Li L, Wage R, Ye X, Slabaugh G, Mohiaddin R, Wong T, Keegan J, Firmin Det al., 2018, Fully automatic segmentation and objective assessment of atrial scars for longstanding persistent atrial fibrillation patients using late gadolinium-enhanced MRI, Medical Physics, Vol: 45, Pages: 1562-1576, ISSN: 0094-2405

PURPOSE: Atrial fibrillation (AF) is the most common heart rhythm disorder and causes considerable morbidity and mortality, resulting in a large public health burden that is increasing as the population ages. It is associated with atrial fibrosis, the amount and distribution of which can be used to stratify patients and to guide subsequent electrophysiology ablation treatment. Atrial fibrosis may be assessed non-invasively using late gadolinium-enhanced (LGE) magnetic resonance imaging (MRI) where scar tissue is visualised as a region of signal enhancement. However, manual segmentation of the heart chambers and of the atrial scar tissue is time-consuming and subject to inter-operator variability, particularly as image quality in AF is often poor. In this study, we propose a novel fully automatic pipeline to achieve accurate and objective segmentation of the heart (from MRI Roadmap data) and of scar tissue within the heart (from LGE MRI data) acquired in patients with AF. METHODS: Our fully automatic pipeline uniquely combines: (1) a multi-atlas based whole heart segmentation (MA-WHS) to determine the cardiac anatomy from an MRI Roadmap acquisition which is then mapped to LGE MRI, and (2) a super-pixel and supervised learning based approach to delineate the distribution and extent of atrial scarring in LGE MRI. We compared the accuracy of the automatic analysis to manual ground-truth segmentations in 37 patients with persistent long standing AF. RESULTS: Both our MA-WHS and atrial scarring segmentations showed accurate delineations of cardiac anatomy (mean Dice = 89%) and atrial scarring (mean Dice = 79%) respectively compared to the established ground truth from manual segmentation. In addition, compared to the ground truth, we obtained 88% segmentation accuracy, with 90% sensitivity and 79% specificity. Receiver operating characteristic analysis achieved an average area under the curve of 0.91. CONCLUSION: Compared with previously studied methods with manual interv

Journal article

Scott AD, Nielles-Vallespin S, Ferreira P, Khalique Z, Gatehouse P, Kilner P, Pennell D, Firmin Det al., 2018, An in-vivo comparison of stimulated-echo and motion compensated spin-echo sequences for 3T diffusion tensor cardiovascular magnetic resonance at multiple cardiac phases, Journal of Cardiovascular Magnetic Resonance, Vol: 20, ISSN: 1097-6647

BackgroundStimulated-echo (STEAM) and, more recently, motion-compensated spin-echo (M2-SE) techniques have been used for in-vivo diffusion tensor cardiovascular magnetic resonance (DT-CMR) assessment of cardiac microstructure. The two techniques differ in the length scales of diffusion interrogated, their signal-to-noise ratio efficiency and sensitivity to both motion and strain. Previous comparisons of the techniques have used high performance gradients at 1.5 T in a single cardiac phase. However, recent work using STEAM has demonstrated novel findings of microscopic dysfunction in cardiomyopathy patients, when DT-CMR was performed at multiple cardiac phases. We compare STEAM and M2-SE using a clinical 3 T scanner in three potentially clinically interesting cardiac phases.MethodsBreath hold mid-ventricular short-axis DT-CMR was performed in 15 subjects using M2-SE and STEAM at end-systole, systolic sweet-spot and diastasis. Success was defined by ≥50% of the myocardium demonstrating normal helix angles. From successful acquisitions DT-CMR results relating to tensor orientation, size and shape were compared between sequences and cardiac phases using non-parametric statistics. Strain information was obtained using cine spiral displacement encoding with stimulated echoes for comparison with DT-CMR results.ResultsAcquisitions were successful in 98% of STEAM and 76% of M2-SE cases and visual helix angle (HA) map scores were higher for STEAM at the sweet-spot and diastasis. There were significant differences between sequences (p < 0.05) in mean diffusivity (MD), fractional anisotropy (FA), tensor mode, transmural HA gradient and absolute second eigenvector angle (E2A). Differences in E2A between systole and diastole correlated with peak radial strain for both sequences (p ≤ 0.01).ConclusionM2-SE and STEAM can be performed equally well at peak systole at 3 T using standard gradients, but at the sweet-spot and diastole STEAM is more rel

Journal article

Khalique Z, Ferreira PF, Scott AD, Nielles-Vallespin S, Firmin DN, Pennell DJet al., 2017, Diffusion tensor CMR in situs inversus: insights into the deranged microstructure and how it affects cardiac function, Publisher: OXFORD UNIV PRESS, Pages: 449-449, ISSN: 0195-668X

Conference paper

Yang G, Zhuang X, Khan H, Haldar S, Nyktari E, Ye X, Slabaugh G, Wong T, Mohiaddin R, Keegan J, Firmin Det al., 2017, Segmenting atrial fibrosis from late gadolinium-enhanced cardiac MRI by deep-learned features with stacked sparse auto-encoders, MIUA 2017, Publisher: Springer, Pages: 195-206, ISSN: 1865-0929

The late gadolinium-enhanced (LGE) MRI technique is a well-validated method for fibrosis detection in the myocardium. With this technique, the altered wash-in and wash-out contrast agent kinetics in fibrotic and healthy myocardium results in scar tissue being seen with high or enhanced signal relative to normal tissue which is ‘nulled’. Recently, great progress on LGE MRI has resulted in improved visualization of fibrosis in the left atrium (LA). This provides valuable information for treatment planning, image-based procedure guidance and clinical management in patients with atrial fibrillation (AF). Nevertheless, precise and objective atrial fibrosis segmentation (AFS) is required for accurate assessment of AF patients using LGE MRI. This is a very challenging task, not only because of the limited quality and resolution of the LGE MRI images acquired in AF but also due to the thinner wall and unpredictable morphology of the LA. Accurate and reliable segmentation of the anatomical structure of the LA myocardium is a prerequisite for accurate AFS. Most current studies rely on manual segmentation of the anatomical structures, which is very labor-intensive and subject to inter- and intra-observer variability. The subsequent AFS is normally based on unsupervised learning methods, e.g., using thresholding, histogram analysis, clustering and graph-cut based approaches, which have variable accuracy. In this study, we present a fully-automated multi-atlas propagation based whole heart segmentation method to derive the anatomical structure of the LA myocardium and pulmonary veins. This is followed by a supervised deep learning method for AFS. Twenty clinical LGE MRI scans from longstanding persistent AF patients were entered into this study retrospectively. We have demonstrated that our fully automatic method can achieve accurate and reliable AFS compared to manual delineated ground truth.

Conference paper

Ferreira PF, Nielles-Vallespin S, Scott AD, Silva RD, Kilner PJ, Ennis DB, Auger DA, Suever JD, Zhong X, Spottiswoode BS, Pennell DJ, Arai AE, Firmin DNet al., 2017, Evaluation of the impact of strain correction on the orientation of cardiac diffusion tensors with in vivo and ex vivo porcine hearts, Magnetic Resonance in Medicine, Vol: 79, Pages: 2205-2215, ISSN: 0740-3194

PurposeTo evaluate the importance of strain-correcting stimulated echo acquisition mode echo-planar imaging cardiac diffusion tensor imaging.MethodsHealthy pigs (n = 11) were successfully scanned with a 3D cine displacement-encoded imaging with stimulated echoes and a monopolar-stimulated echo-planar imaging diffusion tensor imaging sequence at 3 T during diastasis, peak systole, and strain sweet spots in a midventricular short-axis slice. The same diffusion tensor imaging sequence was repeated ex vivo after arresting the hearts in either a relaxed (KCl-induced) or contracted (BaCl2-induced) state. The displacement-encoded imaging with stimulated echoes data were used to strain-correct the in vivo cardiac diffusion tensor imaging in diastole and systole. The orientation of the primary (helix angles) and secondary (E2A) diffusion eigenvectors was compared with and without strain correction and to the strain-free ex vivo data.ResultsStrain correction reduces systolic E2A significantly when compared without strain correction and ex vivo (median absolute E2A = 34.3° versus E2A = 57.1° (P = 0.01), E2A = 60.5° (P = 0.006), respectively). The systolic distribution of E2A without strain correction is closer to the contracted ex vivo distribution than with strain correction, root mean square deviation of 0.027 versus 0.038.ConclusionsThe current strain-correction model amplifies the contribution of microscopic strain to diffusion resulting in an overcorrection of E2A. Results show that a new model that considers cellular rearrangement is required.

Journal article

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: http://wlsprd.imperial.ac.uk:80/respub/WEB-INF/jsp/search-html.jsp Request URI: /respub/WEB-INF/jsp/search-html.jsp Query String: respub-action=search.html&id=00155561&limit=30&person=true