Search or filter publications

Filter by type:

Filter by publication type

Filter by year:



  • Showing results for:
  • Reset all filters

Search results

  • 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
    Mo Y, Liu F, McIlwraith D, Yang G, Zhang J, He T, Guo Yet al., 2018,

    The deep Poincare map: A novel approach for left ventricle segmentation

    , 21st International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 561-568, ISSN: 0302-9743

    Precise segmentation of the left ventricle (LV) within cardiac MRI images is a prerequisite for the quantitative measurement of heart function. However, this task is challenging due to the limited availability of labeled data and motion artifacts from cardiac imaging. In this work, we present an iterative segmentation algorithm for LV delineation. By coupling deep learning with a novel dynamic-based labeling scheme, we present a new methodology where a policy model is learned to guide an agent to travel over the image, tracing out a boundary of the ROI – using the magnitude difference of the Poincaré map as a stopping criterion. Our method is evaluated on two datasets, namely the Sunnybrook Cardiac Dataset (SCD) and data from the STACOM 2011 LV segmentation challenge. Our method outperforms the previous research over many metrics. In order to demonstrate the transferability of our method we present encouraging results over the STACOM 2011 data, when using a model trained on the SCD dataset.

  • Journal article
    Kraaij T, Kamerling SWA, van Dam LS, Bakker JA, Bajema IM, Page T, Brunini F, Pusey CD, Toes REM, Scherer HU, Rabelink TJ, van Kooten C, Teng YKOet al., 2018,

    Excessive neutrophil extracellular trap formation in ANCA-associated vasculitis is independent of ANCA

    , Kidney International, Vol: 94, Pages: 139-149, ISSN: 0085-2538

    Neutrophil extracellular traps (NETs) are auto-antigenic strands of extracellular DNA covered with myeloperoxidase (MPO) and proteinase3 (PR3) that can be a source for the formation of anti-neutrophil cytoplasmic autoantibodies (ANCAs). The presence of NETs was recently demonstrated in renal tissue of patients with ANCA-associated vasculitis (AAV). NET formation was enhanced in AAV, suggesting that MPO-ANCA could trigger NET formation, supporting a vicious circle placing NETs in the center of AAV pathogenesis. Here we investigated NET formation in 99 patients with AAV by a novel highly sensitive and automated assay. There was a significant excess of ex vivo NET formation in both MPO-ANCA- and PR3-ANCA-positive patients with AAV compared to healthy individuals. Excessive NET formation did not correlate with serum ANCA levels. Likewise, immunoglobulin G depletion had no effect on excessive NET formation in patients with AAV, indicating an ANCA-independent process. Next, we explored the relation of excessive NET formation to clinical disease in ten patients with AAV and showed that excessive NET formation was predominantly found during active disease, more so than during remission. Excessive NET formation was found in patients with AAV hospitalized for disease relapse but not during severe infection. Thus, excessive NET formation in AAV is independent of ANCA, and an excess of ex vivo NET formation was related to active clinical disease in patients with AAV and a marker of autoimmunity rather than infection.

  • 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.

  • Conference paper
    Prendecki M, Bhatt T, Dudhiya F, Tam F, Pusey C, McAdoo Set al., 2018,


    , 55th Congress of the European-Renal-Association (ERA) and European-Dialysis-and-Transplantation-Association (EDTA), Publisher: OXFORD UNIV PRESS, ISSN: 0931-0509
  • Journal article
    Alrashed F, Calay D, Lang M, Thornton C, Bauer A, Kiprianos AP, Haskard DO, Seneviratne AN, Boyle J, Schonthal AH, Wheeler-Jones CP, Mason JCet al., 2018,

    Celecoxib exerts protective effects in the vascular endothelium via COX-2-independent activation of AMPK-CREB-Nrf2 signalling

    , Scientific Reports, Vol: 8, ISSN: 2045-2322

    Although concern remains about the athero-thrombotic risk posed by cyclo-oxygenase (COX)-2-selective inhibitors, recent data implicates rofecoxib, while celecoxib appears equivalent to NSAIDs naproxen and ibuprofen. We investigated the hypothesis that celecoxib activates AMP kinase (AMPK) signalling to enhance vascular endothelial protection. In human arterial and venous endothelial cells (EC), and in contrast to ibuprofen and naproxen, celecoxib induced the protective protein heme oxygenase-1 (HO-1). Celecoxib derivative 2,5-dimethyl-celecoxib (DMC) which lacks COX-2 inhibition also upregulated HO-1, implicating a COX-2-independent mechanism. Celecoxib activated AMPKα(Thr172) and CREB-1(Ser133) phosphorylation leading to Nrf2 nuclear translocation. Importantly, these responses were not reproduced by ibuprofen or naproxen, while AMPKα silencing abrogated celecoxib-mediated CREB and Nrf2 activation. Moreover, celecoxib induced H-ferritin via the same pathway, and increased HO-1 and H-ferritin in the aortic endothelium of mice fed celecoxib (1000 ppm) or control chow. Functionally, celecoxib inhibited TNF-α-induced NF-κB p65(Ser536) phosphorylation by activating AMPK. This attenuated VCAM-1 upregulation via induction of HO-1, a response reproduced by DMC but not ibuprofen or naproxen. Similarly, celecoxib prevented IL-1β-mediated induction of IL-6. Celecoxib enhances vascular protection via AMPK-CREB-Nrf2 signalling, a mechanism which may mitigate cardiovascular risk in patients prescribed celecoxib. Understanding NSAID heterogeneity and COX-2-independent signalling will ultimately lead to safer anti-inflammatory drugs.

  • Journal article
    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 long-standing persistent atrial fibrillation patients using late gadolinium-enhanced MRI.

    , Med Phys, Vol: 45, Pages: 1562-1576

    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 noninvasively using late gadolinium-enhanced (LGE) magnetic resonance imaging (MRI) where scar tissue is visualized 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 interoperator 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: (a) a multiatlas-based whole heart segmentation (MA-WHS) to determine the cardiac anatomy from an MRI Roadmap acquisition which is then mapped to LGE MRI, and (b) 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 interve

  • Journal article
    Prendecki M, Martin L, Tanna A, Antonelou M, Pusey CDet al., 2018,

    Increased prevalence of thyroid disease in patients with antineutrophil cytoplasmic antibodies-associated vasculitis

    , Journal of Rheumatology, Vol: 45, ISSN: 0315-162X

    OBJECTIVE: Antineutrophil cytoplasmic antibodies (ANCA)-associated vasculitis (AAV) has been linked with thyroid disease as a result of antithyroid medications. We assessed the prevalence of thyroid disease in our patients with AAV. METHODS: Clinical records of 279 patients with AAV diagnosed between 1991 and 2014 were analyzed. RESULTS: Thyroid disease was identified in 21.5% of patients, but only 2 had previously received propylthiouracil. There was a greater proportion of female patients, patients with antimyeloperoxidase antibodies, and patients with renal disease in the group with thyroid disease. CONCLUSION: Our data show a higher prevalence of thyroid disease in patients with AAV than the general population. This was not attributable to antithyroid drugs.

  • Conference paper
    van Daalen EE, Jennette JC, McAdoo SP, Pusey CD, Alba MA, Poulton CJ, Wolterbeek R, Nguyen TQ, Goldschmeding R, Alchi B, Griffiths M, de Zoysa JR, Vincent B, Bruijn JA, Bajema IMet al., 2018,

    Predicting Outcome in Patients with Anti-GBM Glomerulonephritis

    , 18th International Vasculitis and ANCA Workshop, Publisher: AMER SOC NEPHROLOGY, Pages: 63-72, ISSN: 1555-9041
  • Journal article
    McAdoo SP, Medjeral-Thomas N, Gopaluni S, Tanna A, Mansfield N, Galliford J, Griffith M, Levy J, Cairns T, Jayne D, Salama A, Pusey Cet al.,

    Long-term Follow-up of a Combined Rituximab and Cyclophosphamide Regimen in Renal ANCA-associated Vasculitis

    , Nephrology Dialysis Transplantation, ISSN: 0931-0509

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=1161&limit=10&page=16&respub-action=search.html Current Millis: 1718643186026 Current Time: Mon Jun 17 17:53:06 BST 2024