Search or filter publications

Filter by type:

Filter by publication type

Filter by year:



  • Showing results for:
  • Reset all filters

Search results

  • Conference paper
    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.

  • Journal article
    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 D, 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
    Oliveira V, Singhvi DP, Montaldo P, Lally PJ, Mendoza J, Manerkar S, Shankaran S, Thayyil Set al., 2017,

    Therapeutic hypothermia in mild neonatal encephalopathy: a national survey of practice in the UK

    , Archives of Disease in Childhood. Fetal and Neonatal Edition, Vol: 103, Pages: F388-F390, ISSN: 1359-2998

    Although major cooling trials (and subsequent guidelines) excluded babies with mild encephalopathy, anecdotal evidence suggests that cooling is often offered to these infants. We report a national survey on current cooling practices for babies with mild encephalopathy in the UK. From 74 neonatal units contacted, 68 were cooling centres. We received 54 responses (79%) and included 48 (five excluded due to incomplete data and one found later not to offer cooling). Of these, 36 centres (75%) offered cooling to infants with mild encephalopathy. Although most of the participating units reported targeting 33-34°C core temperature, seven (19%) considered initiating cooling beyond 6 hours of age and 13 (36%) discontinued cooling prior to 72 hours. Babies were ventilated for cooling in two (6%) units and 13 (36%) sedated all cooled babies. Enteral feeding was withheld in 15 (42%) units and reduced below 25% of requirements in eight (22%) units. MRI and neurodevelopmental outcome evaluation were offered to all cooled babies in 29 (80%) and 27 (75%) units, respectively. Further research is necessary to ensure optimal neuroprotection in mild encephalopathy.

  • Journal article
    Thayyil S, Oliveira V, Lally PJ, Swamy R, Bassett P, Chandrasekaran M, Mondkar J, Mangalabharathi S, Benkappa N, Seeralar A, Shahidullah M, Montaldo P, Herberg J, Manerkar S, Kumaraswami K, Kamalaratnam C, Prakash V, Chandramohan R, Bandya P, Mannan MA, Rodrigo R, Nair M, Ramji S, Shankaran S, HELIX Trial groupet al., 2017,

    Hypothermia for encephalopathy in low and middle-income countries (HELIX): study protocol for a randomised controlled trial.

    , Trials, Vol: 18, ISSN: 1745-6215

    BACKGROUND: Therapeutic hypothermia reduces death and disability after moderate or severe neonatal encephalopathy in high-income countries and is used as standard therapy in these settings. However, the safety and efficacy of cooling therapy in low- and middle-income countries (LMICs), where 99% of the disease burden occurs, remains unclear. We will examine whether whole body cooling reduces death or neurodisability at 18-22 months after neonatal encephalopathy, in LMICs. METHODS: We will randomly allocate 408 term or near-term babies (aged ≤ 6 h) with moderate or severe neonatal encephalopathy admitted to public sector neonatal units in LMIC countries (India, Bangladesh or Sri Lanka), to either usual care alone or whole-body cooling with usual care. Babies allocated to the cooling arm will have core body temperature maintained at 33.5 °C using a servo-controlled cooling device for 72 h, followed by re-warming at 0.5 °C per hour. All babies will have detailed infection screening at the time of recruitment and 3 Telsa cerebral magnetic resonance imaging and spectroscopy at 1-2 weeks after birth. Our primary endpoint is death or moderate or severe disability at the age of 18 months. DISCUSSION: Upon completion, HELIX will be the largest cooling trial in neonatal encephalopathy and will provide a definitive answer regarding the safety and efficacy of cooling therapy for neonatal encephalopathy in LMICs. The trial will also provide important data about the influence of co-existent perinatal infection on the efficacy of hypothermic neuroprotection. TRIAL REGISTRATION:, NCT02387385 . Registered on 27 February 2015.

  • Conference paper
    Ghajari M, Hellyer PJ, Sharp DJ, 2017,

    Predicting the location of chronic traumatic encephalopathy pathology

    , 2017 IRCOBI Conference, Publisher: International Research Council on Biomechanics of Injury (IRCOBI), Pages: 699-700, ISSN: 2235-3151

    Chronic traumatic encephalopathy (CTE) is a neurodegenerative disease linked to head impacts. Its distinctive neuropathologic feature is deposition of tau proteins in sulcal depths and in perivascular regions. Previous work has investigated pathological and clinical features of CTE, and here the authors report recent work on exploring the link between strain and strain rate distribution within the brain and location of CTE pathology. The authors used a high fidelity finite element (FE) model of traumatic brain injury (TBI) to test the hypothesis that strain and strain rate produced by head impacts are greatest in sulci, where neuropathology is prominently seen in CTE. The authors also analyzed diffusion tensor imaging (DTI) data from a large cohort of TBI patients to provide converging evidence from empirical neuroimaging data for the model’s prediction.

  • Conference paper
    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

    , International Conference on Medical Image Computing and Computer-Assisted Intervention - MICCAI 2017, Publisher: Springer Verlag, Pages: 253-260, ISSN: 0302-9743

    Training a fully convolutional network for pixel-wise (or voxel-wise) image segmentation normally requires a large number of training images with corresponding ground truth label maps. However, it is a challenge to obtain such a large training set in the medical imaging domain, where expert annotations are time-consuming and difficult to obtain. In this paper, we propose a semi-supervised learning approach, in which a segmentation network is trained from both labelled and unlabelled data. The network parameters and the segmentations for the unlabelled data are alternately updated. We evaluate the method for short-axis cardiac MR image segmentation and it has demonstrated a high performance, outperforming a baseline supervised method. The mean Dice overlap metric is 0.92 for the left ventricular cavity, 0.85 for the myocardium and 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 speed is substantially faster.

  • Journal article
    Datta G, Colasanti A, Kalk N, Owen DR, Scott G, Rabiner EI, Gunn R, Lingford-Hughes A, Malik O, Ciccarelli O, Nicholas R, Nie L, Battaglini M, De Stefano N, Matthews Pet al., 2017,

    [(11)C]PBR28 or [(18)F]PBR111 detect white matter inflammatory heterogeneity in multiple sclerosis

    , Journal of Nuclear Medicine, Vol: 58, Pages: 1477-1482, ISSN: 1535-5667

    Objective: To assess microglial activation in lesions and in normal appearing white matter of multiple sclerosis (MS) patients using positron emission tomography (PET). Methods: 34 MS patients (7 with secondary progressive MS (SPMS), 27 with relapsing remitting MS (RRMS)) and 30 healthy volunteers, genetically stratified for translocator protein (TSPO), binding status underwent PET scanning with TSPO radioligands ((11)C-PBR28 or (18)F-PBR111). Regional TSPO availability was measured as a distribution volume ratio (DVR) relative to the caudate (a pseudo-reference region). White matter lesions (WML) were classified as "active" (DVR highest in the lesion), "peripherally active" (peri-lesional DVR highest), "inactive" (DVR highest in surrounding normal appearing white matter, NAWM) or "undifferentiated" (similar DVR across lesion, peri-lesional and NAWM volumes). Results: The mean DVR in NAWM of patients was greater than that of the healthy volunteer white matter for both radioligands. Uptake for individual WML in patients was heterogeneous, but the median WML DVR and NAWM DVR for individual patients were strongly correlated (ρ = 0.94, P = 4x10-11). A higher proportion of lesions were inactive in patients with SPMS (35 %) than RRMS (23 %), but active lesions were found in all patients, including those on highly efficacious treatments. Conclusion: TSPO radioligand uptake was increased in brains of MS patients relative to healthy controls with two TSPO radiotracers. WML showed heterogeneous patterns of uptake. Active lesions were found in patients with both RRMS and SPMS. Their independent prognostic significance needs further investigation.

  • Journal article
    Tan CL, Alavi SA, Baldeweg SE, Belli A, Carson A, Feeney C, Goldstone AP, Greenwood R, Menon DK, Simpson HL, Toogood AA, Gurnell M, Hutchinson PJet al., 2017,

    The screening and management of pituitary dysfunction following traumatic brain injury in adults: British Neurotrauma Group guidance


    Pituitary dysfunction is a recognised, but potentially underdiagnosed complication of traumatic brain injury (TBI). Post-traumatic hypopituitarism (PTHP) can have major consequences for patients physically, psychologically, emotionally and socially, leading to reduced quality of life, depression and poor rehabilitation outcome. However, studies on the incidence of PTHP have yielded highly variable findings. The risk factors and pathophysiology of this condition are also not yet fully understood. There is currently no national consensus for the screening and detection of PTHP in patients with TBI, with practice likely varying significantly between centres. In view of this, a guidance development group consisting of expert clinicians involved in the care of patients with TBI, including neurosurgeons, neurologists, neurointensivists and endocrinologists, was convened to formulate national guidance with the aim of facilitating consistency and uniformity in the care of patients with TBI, and ensuring timely detection or exclusion of PTHP where appropriate. This article summarises the current literature on PTHP, and sets out guidance for the screening and management of pituitary dysfunction in adult patients with TBI. It is hoped that future research will lead to more definitive recommendations in the form of guidelines.

  • Journal article
    Sliwinska M, Ribeiro Violante I, Wise R, Leech R, Devlin J, Geranmayeh F, Hampshire Aet al., 2017,

    Stimulating Multiple-Demand Cortex Enhances Vocabulary Learning

    , Journal of Neuroscience, Vol: 37, Pages: 7606-7618, ISSN: 1529-2401

    It is well established that domain general networks (DGNs) in the human brain become active when diverse novel skills and behaviors are being learnt. However, their causal role in learning remains to be established. In the present study, we first performed functional magnetic resonance imaging on healthy participants to confirm that DGNs were most active in the initial stages of learning a novel vocabulary, consisting of pronounceable nonwords (pseudowords), each associated with a picture of a real object. We then examined, in healthy participants, whether repetitive transcranial magnetic stimulation of a frontal midline node of the cingulo-opercular DGN affected learning rates during the initial stages of learning. We report that stimulation of this node, but not a control brain region, substantially improved both accuracy and response times during the earliest stage of learning pseudowords-object associations. This stimulation had no effect on the processing of established vocabulary, tested by the accuracy and response times when participants decided whether a real word was accurately paired with a picture of an object. These results provide evidence that non-invasive stimulation to DGN nodes can enhance learning rates, thereby demonstrating their causal role in the learning process. We propose that this causal role makes DGNs candidate targets for experimental therapeutics; for example, in stroke patients with aphasia attempting to reacquire a vocabulary.

  • Journal article
    Gruszka A, Hampshire A, Barker RA, Owen AMet al., 2017,

    Normal aging and Parkinson's disease are associated with the functional decline of distinct frontal-striatal circuits.

    , Cortex, Vol: 93, Pages: 178-192

    Impaired ability to shift attention between stimuli (i.e. shifting attentional 'set') is a well-established part of the dysexecutive syndrome in Parkinson's Disease (PD), nevertheless cognitive and neural bases of this deficit remain unclear. In this study, an fMRI-optimised variant of a classic paradigm for assessing attentional control (Hampshire and Owen 2006) was used to contrast activity in dissociable executive circuits in early-stage PD patients and controls. The results demonstrated that the neural basis of the executive performance impairments in PD is accompanied by hypoactivation within the striatum, anterior cingulate cortex (vACC), and inferior frontal sulcus (IFS) regions. By contrast, in aging it is associated with hypoactivation of the anterior insula/inferior frontal operculum (AI/FO) and the pre-supplementary motor area (preSMA). Between group behavioural differences were also observed; whereas normally aging individuals exhibited routine-problem solving deficits, PD patients demonstrated more global task learning deficits. These findings concur with recent research demonstrating model-based reinforcement learning deficits in PD and provide evidence that the AI/FO and IFS circuits are differentially impacted by PD and normal aging.

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: Request URI: /respub/WEB-INF/jsp/search-t4-html.jsp Query String: id=1053&limit=10&page=8&respub-action=search.html Current Millis: 1586025495957 Current Time: Sat Apr 04 19:38:15 BST 2020