Imperial College London

ProfessorDanielRueckert

Faculty of EngineeringDepartment of Computing

Head of Department of Computing
 
 
 
//

Contact

 

+44 (0)20 7594 8333d.rueckert Website

 
 
//

Location

 

568Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Publication Type
Year
to

697 results found

Cajanus A, Hall A, Koikkalainen J, Solje E, Tolonen A, Urhemaa T, Liu Y, Haanpaa RM, Hartikainen P, Helisalmi S, Korhonen V, Rueckert D, Hasselbalch S, Waldemar G, Mecocci P, Vanninen R, van Gils M, Soininen H, Lotjonen J, Remes AMet al., 2018, Automatic MRI Quantifying Methods in Behavioral-Variant Frontotemporal Dementia Diagnosis, DEMENTIA AND GERIATRIC COGNITIVE DISORDERS EXTRA, Vol: 8, Pages: 51-59, ISSN: 1664-5464

Journal article

Kuklisova-Murgasova M, Estrin GL, Nunes RG, Malik SJ, Rutherford MA, Rueckert D, Hajnal JVet al., 2018, Distortion Correction in Fetal EPI Using Non-Rigid Registration With a Laplacian Constraint, IEEE TRANSACTIONS ON MEDICAL IMAGING, Vol: 37, Pages: 12-19, ISSN: 0278-0062

Journal article

Sinclair M, Peressutti D, Puyol-Anton E, Bai W, Rivolo S, Webb J, Claridge S, Jackson T, Nordsletten D, Hadjicharalambous M, Kerfoot E, Rinaldi CA, Rueckert D, King APet al., 2018, Myocardial strain computed at multiple spatial scales from tagged magnetic resonance imaging: Estimating cardiac biomarkers for CRT patients, MEDICAL IMAGE ANALYSIS, Vol: 43, Pages: 169-185, ISSN: 1361-8415

Journal article

Balaban G, Halliday BP, Costa CM, Porter B, Bai W, Plank G, Rinaldi CA, Rueckert D, Prasad SK, Bishop MJet al., 2018, The Effects of Non-ischemic Fibrosis Texture and Density on Mechanisms of Reentry, 45th Computing in Cardiology Conference (CinC), Publisher: IEEE, ISSN: 2325-8861

Conference paper

Wright R, Khanal B, Gomez A, Skelton E, Matthew J, Hajnal JV, Rueckert D, Schnabel JAet al., 2018, LSTM Spatial Co-transformer Networks for Registration of 3D Fetal US and MR Brain Images, 1st International Workshop on Data Driven Treatment Response Assessment (DATRA) / 3rd International Workshop on Preterm, Perinatal, and Paediatric Image Analysis (PIPPI), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 149-159, ISSN: 0302-9743

Conference paper

Li Y, Alansary A, Cerrolaza JJ, Khanal B, Sinclair M, Matthew J, Gupta C, Knight CL, Kainz B, Rueckert Det al., 2018, Fast Multiple Landmark Localisation Using a Patch-Based Iterative Network., Publisher: Springer, Pages: 563-571

Conference paper

Li Y, Khanal B, Hou B, Alansary A, Cerrolaza JJ, Sinclair M, Matthew J, Gupta C, Knight CL, Kainz B, Rueckert Det al., 2018, Standard Plane Detection in 3D Fetal Ultrasound Using an Iterative Transformation Network., Publisher: Springer, Pages: 392-400

Conference paper

Ktena SI, Parisot S, Ferrante E, Rajchl M, Lee M, Glocker B, Rueckert Det al., 2017, Metric learning with spectral graph convolutions on brain connectivity networks., NeuroImage, Vol: 169, Pages: 431-442, ISSN: 1053-8119

Graph representations are often used to model structured data at an individual or population level and have numerous applications in pattern recognition problems. In the field of neuroscience, where such representations are commonly used to model structural or functional connectivity between a set of brain regions, graphs have proven to be of great importance. This is mainly due to the capability of revealing patterns related to brain development and disease, which were previously unknown. Evaluating similarity between these brain connectivity networks in a manner that accounts for the graph structure and is tailored for a particular application is, however, non-trivial. Most existing methods fail to accommodate the graph structure, discarding information that could be beneficial for further classification or regression analyses based on these similarities. We propose to learn a graph similarity metric using a siamese graph convolutional neural network (s-GCN) in a supervised setting. The proposed framework takes into consideration the graph structure for the evaluation of similarity between a pair of graphs, by employing spectral graph convolutions that allow the generalisation of traditional convolutions to irregular graphs and operates in the graph spectral domain. We apply the proposed model on two datasets: the challenging ABIDE database, which comprises functional MRI data of 403 patients with autism spectrum disorder (ASD) and 468 healthy controls aggregated from multiple acquisition sites, and a set of 2500 subjects from UK Biobank. We demonstrate the performance of the method for the tasks of classification between matching and non-matching graphs, as well as individual subject classification and manifold learning, showing that it leads to significantly improved results compared to traditional methods.

Journal article

Guerrero R, Qin C, Oktay O, Bowles C, Chen L, Joules R, Wolz R, Valdes-Hernandez MC, Dickie DA, Wardlaw J, Rueckert Det al., 2017, White matter hyperintensity and stroke lesion segmentation and differentiation using convolutional neural networks, NeuroImage: Clinical, Vol: 17, Pages: 918-934, ISSN: 2213-1582

White matter hyperintensities (WMH) are a feature of sporadic small vessel disease also frequently observed in magnetic resonance images (MRI) of healthy elderly subjects. The accurate assessment of WMH burden is of crucial importance for epidemiological studies to determine association between WMHs, cognitive and clinical data; their causes, and the effects of new treatments in randomized trials. The manual delineation of WMHs is a very tedious, costly and time consuming process, that needs to be carried out by an expert annotator (e.g. a trained image analyst or radiologist). The problem of WMH delineation is further complicated by the fact that other pathological features (i.e. stroke lesions) often also appear as hyperintense regions. Recently, several automated methods aiming to tackle the challenges of WMH segmentation have been proposed. Most of these methods have been specifically developed to segment WMH in MRI but cannot differentiate between WMHs and strokes. Other methods, capable of distinguishing between different pathologies in brain MRI, are not designed with simultaneous WMH and stroke segmentation in mind. Therefore, a task specific, reliable, fully automated method that can segment and differentiate between these two pathological manifestations on MRI has not yet been fully identified. In this work we propose to use a convolutional neural network (CNN) that is able to segment hyperintensities and differentiate between WMHs and stroke lesions. Specifically, we aim to distinguish between WMH pathologies from those caused by stroke lesions due to either cortical, large or small subcortical infarcts. The proposed fully convolutional CNN architecture, called uResNet, is comprised of an analysis path, that gradually learns low and high level features, followed by a synthesis path, that gradually combines and up-samples the low and high level features into a class likelihood semantic segmentation. Quantitatively, the proposed CNN architecture is shown to

Journal article

Maas AIR, Menon DK, Adelson PD, Andelic N, Bell MJ, Belli A, Bragge P, Brazinova A, Büki A, Chesnut RM, Citerio G, Coburn M, Cooper DJ, Crowder AT, Czeiter E, Czosnyka M, Diaz-Arrastia R, Dreier JP, Duhaime AC, Ercole A, van Essen TA, Feigin VL, Gao G, Giacino J, Gonzalez-Lara LE, Gruen RL, Gupta D, Hartings JA, Hill S, Jiang JY, Ketharanathan N, Kompanje EJO, Lanyon L, Laureys S, Lecky F, Levin H, Lingsma HF, Maegele M, Majdan M, Manley G, Marsteller J, Mascia L, McFadyen C, Mondello S, Newcombe V, Palotie A, Parizel PM, Peul W, Piercy J, Polinder Set al., 2017, Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research, The Lancet Neurology, Vol: 16, Pages: 987-1048, ISSN: 1474-4422

Journal article

Ledig C, Kamnitsas K, Koikkalainen J, Posti JP, Takala RSK, Katila A, Frantzén J, Ala-Seppälä H, Kyllönen A, Maanpää H-R, Tallus J, Lötjönen J, Glocker B, Tenovuo O, Rueckert Det al., 2017, Regional brain morphometry in patients with traumatic brain injury based on acute- and chronic-phase magnetic resonance imaging., PLoS ONE, Vol: 12, ISSN: 1932-6203

Traumatic brain injury (TBI) is caused by a sudden external force and can be very heterogeneous in its manifestation. In this work, we analyse T1-weighted magnetic resonance (MR) brain images that were prospectively acquired from patients who sustained mild to severe TBI. We investigate the potential of a recently proposed automatic segmentation method to support the outcome prediction of TBI. Specifically, we extract meaningful cross-sectional and longitudinal measurements from acute- and chronic-phase MR images. We calculate regional volume and asymmetry features at the acute/subacute stage of the injury (median: 19 days after injury), to predict the disability outcome of 67 patients at the chronic disease stage (median: 229 days after injury). Our results indicate that small structural volumes in the acute stage (e.g. of the hippocampus, accumbens, amygdala) can be strong predictors for unfavourable disease outcome. Further, group differences in atrophy are investigated. We find that patients with unfavourable outcome show increased atrophy. Among patients with severe disability outcome we observed a significantly higher mean reduction of cerebral white matter (3.1%) as compared to patients with low disability outcome (0.7%).

Journal article

Pawlowski N, Ktena SI, Lee MCH, Kainz B, Rueckert D, Glocker B, Rajchl Met al., 2017, DLTK: State of the Art Reference Implementations for Deep Learning on Medical Images

We present DLTK, a toolkit providing baseline implementations for efficientexperimentation with deep learning methods on biomedical images. It builds ontop of TensorFlow and its high modularity and easy-to-use examples allow for alow-threshold access to state-of-the-art implementations for typical medicalimaging problems. A comparison of DLTK's reference implementations of popularnetwork architectures for image segmentation demonstrates new top performanceon the publicly available challenge data "Multi-Atlas Labeling Beyond theCranial Vault". The average test Dice similarity coefficient of $81.5$ exceedsthe previously best performing CNN ($75.7$) and the accuracy of the challengewinning method ($79.0$).

Working paper

Kelly CJ, Makropoulos A, Cordero-Grande L, Hutter J, Price A, Hughes E, Murgasova M, Teixeira RPAG, Steinweg JK, Kulkarni S, Rahman L, Zhang H, Alexander DC, Pushparajah K, Rueckert D, Hajnal JV, Simpson J, Edwards AD, Rutherford MA, Counsell SJet al., 2017, Impaired development of the cerebral cortex in infants with congenital heart disease is correlated to reduced cerebral oxygen delivery, SCIENTIFIC REPORTS, Vol: 7, ISSN: 2045-2322

Journal article

Robinson EC, Garcia K, Glasser MF, Chen Z, Coalson TS, Makropoulos A, Bozek J, Wright R, Schuh A, Webster M, Hutter J, Price A, Grande LC, Hughes E, Tusor N, Bayly PV, Van Essen DC, Smith SM, Edwards AD, Hajnal J, Jenkinson M, Glocker B, Rueckert Det al., 2017, Multimodal surface matching with higher-order smoothness constraints., NeuroImage, Vol: 167, Pages: 453-465, ISSN: 1053-8119

In brain imaging, accurate alignment of cortical surfaces is fundamental to the statistical sensitivity and spatial localisation of group studies; and cortical surface-based alignment has generally been accepted to be superior to volume-based approaches at aligning cortical areas. However, human subjects have considerable variation in cortical folding, and in the location of functional areas relative to these folds. This makes alignment of cortical areas a challenging problem. The Multimodal Surface Matching (MSM) tool is a flexible, spherical registration approach that enables accurate registration of surfaces based on a variety of different features. Using MSM, we have previously shown that driving cross-subject surface alignment, using areal features, such as resting state-networks and myelin maps, improves group task fMRI statistics and map sharpness. However, the initial implementation of MSM's regularisation function did not penalize all forms of surface distortion evenly. In some cases, this allowed peak distortions to exceed neurobiologically plausible limits, unless regularisation strength was increased to a level which prevented the algorithm from fully maximizing surface alignment. Here we propose and implement a new regularisation penalty, derived from physically relevant equations of strain (deformation) energy, and demonstrate that its use leads to improved and more robust alignment of multimodal imaging data. In addition, since spherical warps incorporate projection distortions that are unavoidable when mapping from a convoluted cortical surface to the sphere, we also propose constraints that enforce smooth deformation of cortical anatomies. We test the impact of this approach for longitudinal modelling of cortical development for neonates (born between 31 and 43 weeks of post-menstrual age) and demonstrate that the proposed method increases the biological interpretability of the distortion fields and improves the statistical significance of populatio

Journal article

Schlemper J, Caballero J, Hajnal J, Price A, Rueckert Det al., 2017, A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction, IEEE Transactions on Medical Imaging, Vol: 37, Pages: 491-503, ISSN: 0278-0062

Inspired by recent advances in deep learning, we propose a framework for reconstructing dynamic sequences of 2D cardiac magnetic resonance (MR) images from undersampled data using a deep cascade of convolutional neural networks (CNNs) to accelerate the data acquisition process. In particular, we address the case where data is acquired using aggressive Cartesian undersampling. Firstly, we show that when each 2D image frame is reconstructed independently, the proposed method outperforms state-of-the-art 2D compressed sensing approaches such as dictionary learning-based MR image reconstruction, in terms of reconstruction error and reconstruction speed. Secondly, when reconstructing the frames of the sequences jointly, we demonstrate that CNNs can learn spatio-temporal correlations efficiently by combining convolution and data sharing approaches. We show that the proposed method consistently outperforms state-of-the-art methods and is capable of preserving anatomical structure more faithfully up to 11-fold undersampling. Moreover, reconstruction is very fast: each complete dynamic sequence can be reconstructed in less than 10s and, for the 2D case, each image frame can be reconstructed in 23ms, enabling real-time applications.

Journal article

Sinclair M, Bai W, Puyol-Antón E, Oktay O, Rueckert D, King APet al., 2017, Fully automated segmentation-based respiratory motion correction of multiplanar cardiac magnetic resonance images for large-scale datasets, International Conference On Medical Image Computing & Computer Assisted Intervention, Pages: 332-340, ISSN: 0302-9743

© Springer International Publishing AG 2017. Cardiac magnetic resonance (CMR) can be used for quantitative analysis of heart function. However, CMR imaging typically involves acquiring 2D image planes during separate breath-holds, often resulting in misalignment of the heart between image planes in 3D. Accurate quantitative analysis requires a robust 3D reconstruction of the heart from CMR images, which is adversely affected by such motion artifacts. Therefore, we propose a fully automated method for motion correction of CMR planes using segmentations produced by fully convolutional neural networks (FCNs). FCNs are trained on 100 UK Biobank subjects to produce short-axis and long-axis segmentations, which are subsequently used in an iterative registration algorithm for correcting breath-hold induced motion artifacts. We demonstrate significant improvements in motion-correction over image-based registration, with strong correspondence to results obtained using manual segmentations. We also deploy our automatic method on 9,353 subjects in the UK Biobank database, demonstrating significant improvements in 3D plane alignment.

Conference paper

Oktay O, Ferrante E, Kamnitsas K, Heinrich M, Bai W, Caballero J, Cook S, de Marvao A, Dawes T, O'Regan D, Kainz B, Glocker B, Rueckert Det al., 2017, Anatomically Constrained Neural Networks (ACNN): application to cardiac image enhancement and segmentation, IEEE Transactions on Medical Imaging, Vol: 37, Pages: 384-395, ISSN: 0278-0062

Incorporation of prior knowledge about organ shape and location is key to improve performance of image analysis approaches. In particular, priors can be useful in cases where images are corrupted and contain artefacts due to limitations in image acquisition. The highly constrained nature of anatomical objects can be well captured with learning based techniques. However, in most recent and promising techniques such as CNN based segmentation it is not obvious how to incorporate such prior knowledge. State-of-the-art methods operate as pixel-wise classifiers where the training objectives do not incorporate the structure and inter-dependencies of the output. To overcome this limitation, we propose a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end. The new framework encourages models to follow the global anatomical properties of the underlying anatomy (e.g. shape, label structure) via learnt non-linear representations of the shape. We show that the proposed approach can be easily adapted to different analysis tasks (e.g. image enhancement, segmentation) and improve the prediction accuracy of the state-of-the-art models. The applicability of our approach is shown on multi-modal cardiac datasets and public benchmarks. Additionally, we demonstrate how the learnt deep models of 3D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.

Journal article

Cerrolaza JJ, Oktay O, Gomez A, Matthew J, Knight C, Kainz B, Rueckert Det al., 2017, Fetal skull segmentation in 3D ultrasound via structured geodesic random forest, Fetal, Infant and Ophthalmic Medical Image Analysis: International Workshop, FIFI 2017, and 4th International Workshop, OMIA 2017, Held in Conjunction with MICCAI 2017, Pages: 25-32, ISSN: 0302-9743

© Springer International Publishing AG 2017. Ultrasound is the primary imaging method for prenatal screening and diagnosis of fetal anomalies. Thanks to its non-invasive and non-ionizing properties, ultrasound allows quick, safe and detailed evaluation of the unborn baby, including the estimation of the gestational age, brain and cranium development. However, the accuracy of traditional 2D fetal biometrics is dependent on operator expertise and subjectivity in 2D plane finding and manual marking. 3D ultrasound has the potential to reduce the operator dependence. In this paper, we propose a new random forest-based segmentation framework for fetal 3D ultrasound volumes, able to efficiently integrate semantic and structural information in the classification process. We introduce a new semantic features space able to encode spatial context via generalized geodesic distance transform. Unlike alternative auto-context approaches, this new set of features is efficiently integrated into the same forest using contextual trees. Finally, we use a new structured labels space as alternative to the traditional atomic class labels, able to capture morphological variability of the target organ. Here, we show the potential of this new general framework segmenting the skull in 3D fetal ultrasound volumes, significantly outperforming alternative random forest-based approaches.

Conference paper

Bowles, Qin C, Guerrero R, Gunn R, Hammers A, Dickie D, Valdes Hernandez M, Wardlaw J, Rueckert Det al., 2017, Brain Lesion Segmentation through Image Synthesis and Outlier Detection, NeuroImage: Clinical, Vol: 16, Pages: 643-658, ISSN: 2213-1582

Cerebral small vessel disease (SVD) can manifest in a number of ways. Many of these result in hyperintense regions visible on T2-weighted magnetic resonance (MR) images. The automatic segmentation of these lesions has been the focus of many studies. However, previous methods tended to be limited to certain types of pathology, as a consequence of either restricting the search to the white matter, or by training on an individual pathology. Here we present an unsupervised abnormality detection method which is able to detect abnormally hyperintense regions on FLAIR regardless of the underlying pathology or location. The method uses a combination of image synthesis, Gaussian mixture models and one class support vector machines, and needs only be trained on healthy tissue. We evaluate our method by comparing segmentation results from 127 subjects with SVD with three established methods and report significantly superior performance across a number of metrics.

Journal article

Hou B, Alansary A, McDonagh S, Davidson A, Rutherford M, Hajnal J, Rueckert D, Glocker B, Kainz Bet al., 2017, Predicting slice-to-volume transformation in presence of arbitrary subject motion, 20th International Conference on Medical Image Computing and Computer Assisted Intervention 2017, Publisher: Springer, ISSN: 0302-9743

This paper aims to solve a fundamental problem in intensity-based 2D/3D registration, which concerns the limited capture range andneed for very good initialization of state-of-the-art image registrationmethods. We propose a regression approach that learns to predict rota-tion and translations of arbitrary 2D image slices from 3D volumes, withrespect to a learned canonical atlas co-ordinate system. To this end,we utilize Convolutional Neural Networks (CNNs) to learn the highlycomplex regression function that maps 2D image slices into their cor-rect position and orientation in 3D space. Our approach is attractivein challenging imaging scenarios, where significant subject motion com-plicates reconstruction performance of 3D volumes from 2D slice data.We extensively evaluate the effectiveness of our approach quantitativelyon simulated MRI brain data with extreme random motion. We furtherdemonstrate qualitative results on fetal MRI where our method is in-tegrated into a full reconstruction and motion compensation pipeline.With our CNN regression approach we obtain an average prediction er-ror of 7mm on simulated data, and convincing reconstruction quality ofimages of very young fetuses where previous methods fail. We furtherdiscuss applications to Computed Tomography (CT) and X-Ray pro-jections. Our approach is a general solution to the 2D/3D initializationproblem. It is computationally efficient, with prediction times per sliceof a few milliseconds, making it suitable for real-time scenarios.

Conference paper

Garcia KE, Robinson EC, Alexopoulos D, Dierker DL, Glasser MF, Coalson TS, Ortinau CM, Rueckert D, Taber LA, Van Essen DC, Rogers CE, Smyser CD, Bayly PVet al., 2017, Dynamic patterns of cortical expansion during folding of the preterm human brain, Publisher: Cold Spring Harbor Laboratory

<jats:p>During the third trimester of human brain development, the cerebral cortex undergoes dramatic surface expansion and folding. Physical models suggest that relatively rapid growth of the cortical gray matter helps drive this folding, and structural data suggests that growth may vary in both space (by region on the cortical surface) and time. In this study, we propose a new method to estimate local growth from sequential cortical reconstructions. Using anatomically-constrained Multimodal Surface Matching (aMSM), we obtain accurate, physically-guided point correspondence between younger and older cortical reconstructions of the same individual. From each pair of surfaces, we calculate continuous, smooth maps of cortical expansion with unprecedented precision. By considering 30 preterm infants scanned 2-4 times during the period of rapid cortical expansion (28 to 38 weeks postmenstrual age), we observe significant regional differences in growth across the cortical surface that are consistent with patterns of active folding. Furthermore, these growth patterns shift over the course of development, with non-injured subjects following a highly consistent trajectory. This information provides a detailed picture of dynamic changes in cortical growth, connecting what is known about patterns of development at the microscopic (cellular) and macroscopic (folding) scales. Since our method provides specific growth maps for individual brains, we are also able to detect alterations due to injury. This fully-automated surface analysis, based on tools freely available to the brain mapping community, may also serve as a useful approach for future studies of abnormal growth due to genetic disorders, injury, or other environmental variables.</jats:p><jats:sec><jats:title>Significance Statement</jats:title><jats:p>The human brain exhibits complex folding patterns that emerge during the third trimester of fetal development. Minor folds are quasi-ra

Working paper

Kanavati F, Misawa K, Fujiwara M, Mori K, Rueckert D, Glocker Bet al., 2017, Joint Supervoxel Classification Forest for Weakly-Supervised Organ Segmentation, International Workshop on Machine Learning in Medical Imaging (MLMI)

Conference paper

Parisot S, Glocker B, Ktena SI, Arslan S, Schirmer MD, Rueckert Det al., 2017, A flexible graphical model for multi-modal parcellation of the cortex., NeuroImage, Vol: 162, Pages: 226-248, ISSN: 1053-8119

Advances in neuroimaging have provided a tremendous amount of in-vivo information on the brain's organisation. Its anatomy and cortical organisation can be investigated from the point of view of several imaging modalities, many of which have been studied for mapping functionally specialised cortical areas. There is strong evidence that a single modality is not sufficient to fully identify the brain's cortical organisation. Combining multiple modalities in the same parcellation task has the potential to provide more accurate and robust subdivisions of the cortex. Nonetheless, existing brain parcellation methods are typically developed and tested on single modalities using a specific type of information. In this paper, we propose Graph-based Multi-modal Parcellation (GraMPa), an iterative framework designed to handle the large variety of available input modalities to tackle the multi-modal parcellation task. At each iteration, we compute a set of parcellations from different modalities and fuse them based on their local reliabilities. The fused parcellation is used to initialise the next iteration, forcing the parcellations to converge towards a set of mutually informed modality specific parcellations, where correspondences are established. We explore two different multi-modal configurations for group-wise parcellation using resting-state fMRI, diffusion MRI tractography, myelin maps and task fMRI. Quantitative and qualitative results on the Human Connectome Project database show that integrating multi-modal information yields a stronger agreement with well established atlases and more robust connectivity networks that provide a better representation of the population.

Journal article

Cnossen MC, Huijben JA, van der Jagt M, Volovici V, van Essen T, Polinder S, Nelson D, Ercole A, Stocchetti N, Citerio G, Peul WC, Maas AIR, Menon D, Steyerberg EW, Lingsma HF, Adams H, Alessandro M, Allanson J, Amrein K, Andaluz N, Andelic N, Andrea N, Andreassen L, Anke A, Antoni A, Ardon H, Audibert G, Auslands K, Azouvi P, Baciu C, Bacon A, Badenes R, Baglin T, Bartels R, Barzó P, Bauerfeind U, Beer R, Belda FJ, Bellander BM, Belli A, Bellier R, Benali H, Benard T, Berardino M, Beretta L, Beynon C, Bilotta F, Binder H, Biqiri E, Blaabjerg M, Lund SB, Bouzat P, Bragge P, Brazinova A, Brehar F, Brorsson C, Buki A, Bullinger M, Bucková V, Calappi E, Cameron P, Carbayo LG, Carise E, Carpenter K, Castaño-León AM, Causin F, Chevallard G, Chieregato A, Cooper M, Cnossen M, Coburn M, Coles J, Cooper JD, Correia M, Covic A, Curry N, Czeiter E, Czosnyka M, Dahyot-Fizelier C, Damas F, Damas P, Dawes H, De Keyser V, Corte FD, Depreitere B, Ding S, Dippel D, Dizdarevic K, Dulière GL, Dzeko A, Eapen G, Engemann H, Ercole Aet al., 2017, Variation in monitoring and treatment policies for intracranial hypertension in traumatic brain injury: A survey in 66 neurotrauma centers participating in the CENTER-TBI study, Critical Care, Vol: 21, ISSN: 1364-8535

Background: No definitive evidence exists on how intracranial hypertension should be treated in patients with traumatic brain injury (TBI). It is therefore likely that centers and practitioners individually balance potential benefits and risks of different intracranial pressure (ICP) management strategies, resulting in practice variation. The aim of this study was to examine variation in monitoring and treatment policies for intracranial hypertension in patients with TBI. Methods: A 29-item survey on ICP monitoring and treatment was developed on the basis of literature and expert opinion, and it was pilot-tested in 16 centers. The questionnaire was sent to 68 neurotrauma centers participating in the Collaborative European Neurotrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) study. Results: The survey was completed by 66 centers (97% response rate). Centers were mainly academic hospitals (n=60, 91%) and designated level I trauma centers (n=44, 67%). The Brain Trauma Foundation guidelines were used in 49 (74%) centers. Approximately 90% of the participants (n=58) indicated placing an ICP monitor in patients with severe TBI and computed tomographic abnormalities. There was no consensus on other indications or on peri-insertion precautions. We found wide variation in the use of first- and second-tier treatments for elevated ICP. Approximately half of the centers were classified as using a relatively aggressive approach to ICP monitoring and treatment (n=32, 48%), whereas the others were considered more conservative (n=34, 52%). Conclusions: Substantial variation was found regarding monitoring and treatment policies in patients with TBI and intracranial hypertension. The results of this survey indicate a lack of consensus between European neurotrauma centers and provide an opportunity and necessity for comparative effectiveness research.

Journal article

Robinson R, Valindria V, Bai W, Suzuki H, Matthews P, Page C, Rueckert D, Glocker Bet al., 2017, Automatic quality control of cardiac MRI segmentation in large-scale population imaging, Medical Image Computing and Computer Assisted Intervention, Publisher: Springer, ISSN: 0302-9743

The trend towards large-scale studies including populationimaging poses new challenges in terms of quality control (QC). This is aparticular issue when automatic processing tools such as image segmenta-tion methods are employed to derive quantitative measures or biomarkersfor further analyses. Manual inspection and visual QC of each segmen-tation result is not feasible at large scale. However, it is important tobe able to detect when an automatic method fails to avoid inclusionof wrong measurements into subsequent analyses which could otherwiselead to incorrect conclusions. To overcome this challenge, we explorean approach for predicting segmentation quality based on reverse clas-sification accuracy, which enables us to discriminate between successfuland failed cases. We validate this approach on a large cohort of cardiacMRI for which manual QC scores were available. Our results on 7,425cases demonstrate the potential for fully automatic QC in the context oflarge-scale population imaging such as the UK Biobank Imaging Study.

Conference paper

Parisot S, Ktena SI, Ferrante E, Lee M, Moreno RG, Glocker B, Rueckert Det al., 2017, Spectral Graph Convolutions for Population-based Disease Prediction, Medical Image Computing and Computer Assisted Intervention, Publisher: Springer, ISSN: 0302-9743

Exploiting the wealth of imaging and non-imaging information for diseaseprediction tasks requires models capable of representing, at the same time,individual features as well as data associations between subjects frompotentially large populations. Graphs provide a natural framework for suchtasks, yet previous graph-based approaches focus on pairwise similaritieswithout modelling the subjects' individual characteristics and features. On theother hand, relying solely on subject-specific imaging feature vectors fails tomodel the interaction and similarity between subjects, which can reduceperformance. In this paper, we introduce the novel concept of GraphConvolutional Networks (GCN) for brain analysis in populations, combiningimaging and non-imaging data. We represent populations as a sparse graph whereits vertices are associated with image-based feature vectors and the edgesencode phenotypic information. This structure was used to train a GCN model onpartially labelled graphs, aiming to infer the classes of unlabelled nodes fromthe node features and pairwise associations between subjects. We demonstratethe potential of the method on the challenging ADNI and ABIDE databases, as aproof of concept of the benefit from integrating contextual information inclassification tasks. This has a clear impact on the quality of thepredictions, leading to 69.5% accuracy for ABIDE (outperforming the currentstate of the art of 66.8%) and 77% for ADNI for prediction of MCI conversion,significantly outperforming standard linear classifiers where only individualfeatures are considered.

Conference paper

Biffi C, Simoes Monteiro de Marvao A, Attard M, Dawes T, Whiffin N, Bai W, Shi W, Francis C, Meyer H, Buchan R, Cook S, Rueckert D, O'Regan DPet al., 2017, Three-dimensional Cardiovascular Imaging-Genetics: A Mass Univariate Framework, Bioinformatics, ISSN: 1367-4803

Motivation: Left ventricular (LV) hypertrophy is a strong predictor of cardiovascular outcomes, but its genetic regulation remains largely unexplained. Conventional phenotyping relies on manual calculation of LV mass and wall thickness, but advanced cardiac image analysis presents an opportunity for highthroughput mapping of genotype-phenotype associations in three dimensions (3D).Results: High-resolution cardiac magnetic resonance images were automatically segmented in 1,124 healthy volunteers to create a 3D shape model of the heart. Mass univariate regression was used to plot a 3D effect-size map for the association between wall thickness and a set of predictors at each vertex in the mesh. The vertices where a significant effect exists were determined by applying threshold-free cluster enhancement to boost areas of signal with spatial contiguity. Experiments on simulated phenotypic signals and SNP replication show that this approach offers a substantial gain in statistical power for cardiac genotype-phenotype associations while providing good control of the false discovery rate. This framework models the effects of genetic variation throughout the heart and can be automatically applied to large population cohorts.Availability: The proposed approach has been coded in an R package freely available at https://doi.org/10.5281/zenodo.834610 together with the clinical data used in this work.

Journal article

Bai W, Oktay O, Sinclair M, Suzuki H, Rajchl M, Tarroni G, Glocker B, King A, Matthews P, Rueckert Det al., 2017, Semi-supervised learning for network-based cardiac MR image segmentation, Medical Image Computing and Computer Assisted Intervention, Publisher: Springer, ISSN: 0302-9743

Training a fully convolutional network for pixel-wise (or voxel-wise) image segmentation normally requires a large number of trainingimages with corresponding ground truth label maps. However, it is a chal-lenge to obtain such a large training set in the medical imaging domain,where expert annotations are time-consuming and difficult to obtain. Inthis paper, we propose a semi-supervised learning approach, in which asegmentation network is trained from both labelled and unlabelled data.The network parameters and the segmentations for the unlabelled dataare alternately updated. We evaluate the method for short-axis cardiacMR image segmentation and it has demonstrated a high performance,outperforming a baseline supervised method. The mean Dice overlapmetric is 0.92 for the left ventricular cavity, 0.85 for the myocardiumand 0.89 for the right ventricular cavity. It also outperforms a state-of-the-art multi-atlas segmentation method by a large margin and the speedis substantially faster.

Conference paper

Ktena SI, Parisot S, Ferrante E, Rajchl M, Lee M, Glocker B, Rueckert Det al., 2017, Distance metric learning using graph convolutional networks: application to functional brain networks, Medical Image Computing and Computer Assisted Intervention, Publisher: Springer, ISSN: 0302-9743

Evaluating similarity between graphs is of major importance in severalcomputer vision and pattern recognition problems, where graph representationsare often used to model objects or interactions between elements. The choice ofa distance or similarity metric is, however, not trivial and can be highlydependent on the application at hand. In this work, we propose a novel metriclearning method to evaluate distance between graphs that leverages the power ofconvolutional neural networks, while exploiting concepts from spectral graphtheory to allow these operations on irregular graphs. We demonstrate thepotential of our method in the field of connectomics, where neuronal pathwaysor functional connections between brain regions are commonly modelled asgraphs. In this problem, the definition of an appropriate graph similarityfunction is critical to unveil patterns of disruptions associated with certainbrain disorders. Experimental results on the ABIDE dataset show that our methodcan learn a graph similarity metric tailored for a clinical application,improving the performance of a simple k-nn classifier by 11.9% compared to atraditional distance metric.

Conference paper

Alansary A, Rajchl M, McDonagh S, Murgasova M, Damodaram M, Lloyd DFA, Davidson A, Rutherford M, Hajnal JV, Rueckert D, Kainz Bet al., 2017, PVR: Patch-to-Volume Reconstruction for Large Area Motion Correction of Fetal MRI, IEEE Transactions on Medical Imaging, Vol: 36, Pages: 2031-2044, ISSN: 1558-254X

In this paper we present a novel method for the correction of motion artifacts that are present in fetal Magnetic Resonance Imaging (MRI) scans of the whole uterus. Contrary to current slice-to-volume registration (SVR) methods, requiring an inflexible anatomical enclosure of a \emph{single} investigated organ, the proposed patch-to-volume reconstruction (PVR) approach is able to reconstruct a large field of view of non-rigidly deforming structures. It relaxes rigid motion assumptions by introducing a specific amount of redundant information that is exploited with parallelized patch-wise optimization, super-resolution, and automatic outlier rejection. We further describe and provide an efficient parallel implementation of PVR allowing its execution within reasonable time on commercially available graphics processing units (GPU), enabling its use in the clinical practice. We evaluate PVR's computational overhead compared to standard methods and observe improved reconstruction accuracy in the presence of affine motion artifacts compared to conventional SVR in synthetic experiments. Furthermore, we have evaluated our method qualitatively and quantitatively on real fetal MRI data subject to maternal breathing and sudden fetal movements. We evaluate peak-signal-to-noise ratio (PSNR), structural similarity index (SSIM), and cross correlation (CC) with respect to the originally acquired data and provide a method for visual inspection of reconstruction uncertainty. We further evaluate the distance error for selected anatomical landmarks in the fetal head, as well as calculating the mean and maximum displacements resulting from automatic non-rigid registration to a motion-free ground truth image. These experiments demonstrate a successful application of PVR motion compensation to the whole fetal body, uterus and placenta.

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: id=00172041&limit=30&person=true&page=4&respub-action=search.html