Imperial College London

Professor Paul M. Matthews

Faculty of MedicineDepartment of Brain Sciences

Edmond and Lily Safra Chair. Head of Department
 
 
 
//

Contact

 

+44 (0)20 7594 2855p.matthews

 
 
//

Assistant

 

Ms Siobhan Dillon +44 (0)20 7594 2855

 
//

Location

 

E502Burlington DanesHammersmith Campus

//

Summary

 

Publications

Publication Type
Year
to

599 results found

Supratak A, Datta G, Gafson AR, Nicholas R, Guo Y, Matthews PMet al., 2018, Remote monitoring in the home validates clinical gait measures for multiple sclerosis, Frontiers in Neurology, Vol: 9, Pages: 1-9, ISSN: 1664-2295

Background: The timed 25-foot walk (T25FW) is widely used as a clinic performance measure, but has yet to be directly validated against gait speed in the home environment.Objectives: To develop an accurate method for remote assessment of walking speed and to test how predictive the clinic T25FW is for real-life walking.Methods: An AX3-Axivity tri-axial accelerometer was positioned on 32 MS patients (Expanded Disability Status Scale [EDSS] 0–6) in the clinic, who subsequently wore it at home for up to 7 days. Gait speed was calculated from these data using both a model developed with healthy volunteers and individually personalized models generated from a machine learning algorithm.Results: The healthy volunteer model predicted gait speed poorly for more disabled people with MS. However, the accuracy of individually personalized models was high regardless of disability (R-value = 0.98, p-value = 1.85 × 10−22). With the latter, we confirmed that the clinic T25FW is strongly predictive of the maximum sustained gait speed in the home environment (R-value = 0.89, p-value = 4.34 × 10−8).Conclusion: Remote gait monitoring with individually personalized models is accurate for patients with MS. Using these models, we have directly validated the clinical meaningfulness (i.e., predictiveness) of the clinic T25FW for the first time.

Journal article

Gafson A, Kicheol K, Cencioni M, Van Hecke W, Nicholas R, Baranzini S, Matthews Pet al., 2018, Mononuclear cell transcriptome changes associated with dimethyl fumarate in multiple sclerosis., Neurology, Neuroimmunology and Neuroinflammation, Vol: 5, ISSN: 2332-7812

Objective To identify short-term changes in gene expression in peripheral blood mononuclear cells (PBMCs) associated with treatment response to dimethyl fumarate (DMF, Tecfidera) in patients with relapsing-remitting MS (RRMS).Methods Blood samples were collected from 24 patients with RRMS (median Expanded Disability Status Scale score, 2.0; range 1–7) at baseline, 6 weeks, and 15 months after the initiation of treatment with DMF (BG-12; Tecfidera). Seven healthy controls were also recruited, and blood samples were collected over the same time intervals. PBMCs were extracted from blood samples and sequenced using next-generation RNA sequencing. Treatment responders were defined using the composite outcome measure “no evidence of disease activity” (NEDA-4). Time-course and cross-sectional differential expression analyses were performed to identify transcriptomic markers of treatment response.Results Treatment responders (NEDA-4 positive, 8/24) over the 15-month period had 478 differentially expressed genes (DEGs) 6 weeks after the start of treatment. These were enriched for nuclear factor (erythroid-derived 2)-like 2 (Nrf2) and inhibition of nuclear factor κB (NFκB) pathway transcripts. For patients who showed signs of disease activity, there were no DEGs at 6 weeks relative to their (untreated) baseline. Contrasting transcriptomes expressed at 6 weeks with those at 15 months of treatment, 0 and 1,264 DEGs were found in the responder and nonresponder groups, respectively. Transcripts in the nonresponder group (NEDA-4 negative, 18/24) were enriched for T-cell signaling genes.Conclusion Short-term PBMC transcriptome changes reflecting activation of the Nrf2 and inhibition of NFκB pathways distinguish patients who subsequently show a medium-term treatment response with DMF. Relative stabilization of gene expression patterns may accompany treatment-associated suppression of disease activity.

Journal article

Gibson L, Littlejohns T, Adamska L, Garratt S, Doherty N, Wardlaw J, Maskell G, Parker M, Brownsword R, Matthews P, Collins R, Allen N, Sellors J, Sudlow CLM, UK Biobank Imaging Working Groupet al., 2018, Impact of detecting potentially serious incidental findings during multi-modal imaging

Background : There are limited data on the impact of feedback of incidental findings (IFs) from research imaging.  We evaluated the impact of UK Biobank’s protocol for handling potentially serious IFs in a multi-modal imaging study of 100,000 participants (radiographer ‘flagging’ with radiologist confirmation of potentially serious IFs) compared with systematic radiologist review of all images. Methods : Brain, cardiac and body magnetic resonance, and dual-energy x-ray absorptiometry scans from the first 1000 imaged UK Biobank participants were independently assessed for potentially serious IFs using both protocols. We surveyed participants with potentially serious IFs and their GPs up to six months after imaging to determine subsequent clinical assessments, final diagnoses, emotional, financial and work or activity impacts. Results : Compared to systematic radiologist review, radiographer flagging resulted in substantially fewer participants with potentially serious IFs (179/1000 [17.9%] versus 18/1000 [1.8%]) and a higher proportion with serious final diagnoses (21/179 [11.7%] versus 5/18 [27.8%]). Radiographer flagging missed 16/21 serious final diagnoses (i.e., false negatives), while systematic radiologist review generated large numbers of non-serious final diagnoses (158/179) (i.e., false positives). Almost all (90%) participants had further clinical assessment (including invasive procedures in similar numbers with serious and non-serious final diagnoses [11 and 12 respectively]), with additional impact on emotional wellbeing (16.9%), finances (8.9%), and work or activities (5.6%). Conclusions : Compared with systematic radiologist review, radiographer flagging missed some serious diagnoses, but avoided adverse impacts for many participants with non-serious diagnoses. While systematic radiologist review may benefit some participants, UK Biobank’s responsibility to avoid both unnecessary harm to larger numbers of

Working paper

Dawes TJW, Serrani M, Bai W, Cai J, Suzuki H, de Marvao A, Quinlan M, Tokarczuk P, Ostrowski P, Matthews P, Rueckert D, Cook S, Costantino ML, O'Regan Det al., Myocardial trabeculae improve left ventricular function: a combined UK Biobank and computational analysis, GAT Annual Scientific Meeting 2018, Publisher: Association of Anaesthetists of Great Britain and Ireland

Conference paper

Bishop CA, Ricotti V, Sinclair CDJ, Evans MRB, Butler JW, Morrow JM, Hanna MG, Matthews PM, Yousry TA, Muntoni F, Thornton JS, Newbould RD, Janiczek RLet al., 2018, Semi-automated analysis of diaphragmatic motion with dynamic magnetic resonance imaging in healthy controls and non-ambulant subjects with duchenne muscular dystrophy, Frontiers in Neurology, Vol: 9, ISSN: 1664-2295

Subjects with Duchenne Muscular Dystrophy (DMD) suffer from progressive muscle damage leading to diaphragmatic weakness that ultimately requires ventilation. Emerging treatments have generated interest in better characterizing the natural history of respiratory impairment in DMD and responses to therapy. Dynamic (cine) Magnetic Resonance Imaging (MRI) may provide a more sensitive measure of diaphragm function in DMD than the commonly used spirometry. This study presents an analysis pipeline for measuring parameters of diaphragmatic motion from dynamic MRI and its application to investigate MRI measures of respiratory function in both healthy controls and non-ambulant DMD boys. We scanned 13 non-ambulant DMD boys and 10 age-matched healthy male volunteers at baseline, with a subset (n = 10, 10, 8) of the DMD subjects also assessed 3, 6, and 12 months later. Spirometry-derived metrics including forced vital capacity were recorded. The MRI-derived measures included the lung cross-sectional area (CSA), the anterior, central, and posterior lung lengths in the sagittal imaging plane, and the diaphragm length over the time-course of the dynamic MRI. Regression analyses demonstrated strong linear correlations between lung CSA and the length measures over the respiratory cycle, with a reduction of these correlations in DMD, and diaphragmatic motions that contribute less efficiently to changing lung capacity in DMD. MRI measures of pulmonary function were reduced in DMD, controlling for height differences between the groups: at maximal inhalation, the maximum CSA and the total distance of motion of the diaphragm were 45% and 37% smaller. MRI measures of pulmonary function were correlated with spirometry data and showed relationships with disease progression surrogates of age and months non-ambulatory, suggesting that they provide clinically meaningful information. Changes in the MRI measures over 12 months were consistent with weakening of diaphragmatic and inter-costal muscl

Journal article

Scott GPT, Zetterberg H, Jolly A, Cole JH, De Simoni S, Jenkins PO, Feeney C, Owen DR, Lingford-Hughes A, Howes O, Patel MC, Goldstone AP, Gunn RN, Blennow K, Matthews PM, Sharp DJet al., 2017, Minocycline reduces chronic microglial activation after brain trauma but increases neurodegeneration, Brain, Vol: 141, Pages: 459-471, ISSN: 1460-2156

Survivors of a traumatic brain injury can deteriorate years later, developing brain atrophy and dementia. Traumatic brain injury triggers chronic microglial activation, but it is unclear whether this is harmful or beneficial. A successful chronic-phase treatment for traumatic brain injury might be to target microglia. In experimental models, the antibiotic minocycline inhibits microglial activation. We investigated the effect of minocycline on microglial activation and neurodegeneration using PET, MRI, and measurement of the axonal protein neurofilament light in plasma. Microglial activation was assessed using 11C-PBR28 PET. The relationships of microglial activation to measures of brain injury, and the effects of minocycline on disease progression, were assessed using structural and diffusion MRI, plasma neurofilament light, and cognitive assessment. Fifteen patients at least 6 months after a moderate-to-severe traumatic brain injury received either minocycline 100 mg orally twice daily or no drug, for 12 weeks. At baseline, 11C-PBR28 binding in patients was increased compared to controls in cerebral white matter and thalamus, and plasma neurofilament light levels were elevated. MRI measures of white matter damage were highest in areas of greater 11C-PBR28 binding. Minocycline reduced 11C-PBR28 binding (mean Δwhite matter binding = −23.30%, 95% confidence interval −40.9 to −5.64%, P = 0.018), but increased plasma neurofilament light levels. Faster rates of brain atrophy were found in patients with higher baseline neurofilament light levels. In this experimental medicine study, minocycline after traumatic brain injury reduced chronic microglial activation while increasing a marker of neurodegeneration. These findings suggest that microglial activation has a reparative effect in the chronic phase of traumatic brain injury.

Journal article

Stangel M, Kuhlmann T, Matthews PM, Kilpatrick TJet al., 2017, Achievements and obstacles of remyelinating therapies in multiple sclerosis, Nature Reviews Neurology, Vol: 13, Pages: 742-754, ISSN: 1759-4758

Remyelination in the CNS is the natural process of damage repair in demyelinating diseases such as multiple sclerosis (MS). However, remyelination becomes inadequate in many people with MS, which results in axonal degeneration and clinical disability. Enhancement of remyelination is a logical therapeutic goal; nevertheless, all currently licensed therapies for MS are immunomodulatory and do not support remyelination directly. Several molecular pathways have been identified as potential therapeutic targets to induce remyelination, and some of these have now been assessed in proof-of-concept clinical trials. However, trial design faces several obstacles: optimal clinical or paraclinical outcome measures to assess remyelination remain ill-defined, and identification of the ideal timing of therapy is also a crucial issue. In addition, realistic expectations are needed concerning the probable benefits of such therapies. Nevertheless, approaches that enhance remyelination are likely to be protective for axons and so could prevent long-term neurodegeneration. Future MS treatment paradigms, therefore, are likely to comprise a combinatorial approach that involves both immunomodulatory and regenerative treatments.

Journal article

Gibson LM, Littlejohns TJ, Adamska L, Garratt S, Doherty N, UK Biobank Imaging Working Group, Wardlaw JM, Maskell G, Parker M, Brownsword R, Matthews PM, Collins R, Allen NE, Sellors J, Sudlow CLet al., 2017, Impact of detecting potentially serious incidental findings during multi-modal imaging [version 3; peer review: 2 approved, 1 approved with reservations], Wellcome Open Research, Vol: 2, ISSN: 2398-502X

Background: There are limited data on the impact of feedback of incidental findings (IFs) from research imaging.  We evaluated the impact of UK Biobank's protocol for handling potentially serious IFs in a multi-modal imaging study of 100,000 participants (radiographer 'flagging' with radiologist confirmation of potentially serious IFs) compared with systematic radiologist review of all images. Methods: Brain, cardiac and body magnetic resonance, and dual-energy x-ray absorptiometry scans from the first 1000 imaged UK Biobank participants were independently assessed for potentially serious IFs using both protocols. We surveyed participants with potentially serious IFs and their GPs up to six months after imaging to determine subsequent clinical assessments, final diagnoses, emotional, financial and work or activity impacts. Results: Compared to systematic radiologist review, radiographer flagging resulted in substantially fewer participants with potentially serious IFs (179/1000 [17.9%] versus 18/1000 [1.8%]) and a higher proportion with serious final diagnoses (21/179 [11.7%] versus 5/18 [27.8%]). Radiographer flagging missed 16/21 serious final diagnoses (i.e., false negatives), while systematic radiologist review generated large numbers of non-serious final diagnoses (158/179) (i.e., false positives). Almost all (90%) participants had further clinical assessment (including invasive procedures in similar numbers with serious and non-serious final diagnoses [11 and 12 respectively]), with additional impact on emotional wellbeing (16.9%), finances (8.9%), and work or activities (5.6%). Conclusions: Compared with systematic radiologist review, radiographer flagging missed some serious diagnoses, but avoided adverse impacts for many participants with non-serious diagnoses. While systematic radiologist review may benefit some participants, UK Biobank's responsibility to avoid both unnecessary harm to larger numbers of participants and burdening

Journal article

Owen DRJ, Fan J, Campioli E, Venugopal S, Midzak A, Daly E, Harlay A, Issop L, Libri V, Kalogiannopoulou D, Oliver E, Gallego-Colon E, Colasanti A, Huson L, Rabiner EA, Suppiah P, Essagian C, Matthews PM, Papadopoulos Vet al., 2017, TSPO mutations in rats and a human polymorphism impair the rate of steroid synthesis, Biochemical Journal, Vol: 474, Pages: 3985-3999, ISSN: 1470-8728

The 18 kDa translocator protein (TSPO) is a ubiquitous conserved outer mitochondrial membrane protein implicated in numerous cell and tissue functions, including steroid hormone biosynthesis, respiration, cell proliferation, and apoptosis. TSPO binds with high affinity to cholesterol and numerous compounds, is expressed at high levels in steroid-synthesizing tissues, and mediates cholesterol import into mitochondria, which is the rate-limiting step in steroid formation. In humans, the rs6971 polymorphism on the TSPO gene leads to an amino acid substitution in the fifth transmembrane loop of the protein, which is where the cholesterol-binding domain of TSPO is located, and this polymorphism has been associated with anxiety-related disorders. However, recent knockout mouse models have provided inconsistent conclusions of whether TSPO is directly involved in steroid synthesis. In this report, we show that TSPO deletion mutations in rat and its corresponding rs6971 polymorphism in humans alter adrenocorticotropic hormone-induced plasma corticosteroid concentrations. Rat tissues examined show increased cholesteryl ester accumulation, and neurosteroid formation was undetectable in homozygous rats. These results also support a role for TSPO ligands in diseases with steroid-dependent stress and anxiety elements.

Journal article

Suzuki HS, Gao HG, Bai WB, Evangelou EE, Glocker BG, O'regan DO, Elliott PE, Matthews PMMet al., 2017, Abnormal brain white matter microstructure is associated withboth pre-hypertension and hypertension, PLoS ONE, Vol: 12, ISSN: 1932-6203

ObjectivesTo characterize effects of chronically elevated blood pressure on the brain, we tested for brain white matter microstructural differences associated with normotension, pre-hypertension and hypertension in recently available brain magnetic resonance imaging data from 4659 participants without known neurological or psychiatric disease (62.3±7.4 yrs, 47.0% male) in UK Biobank.MethodsFor assessment of white matter microstructure, we used measures derived from neurite orientation dispersion and density imaging (NODDI) including the intracellular volume fraction (an estimate of neurite density) and isotropic volume fraction (an index of the relative extra-cellular water diffusion). To estimate differences associated specifically with blood pressure, we applied propensity score matching based on age, sex, educational level, body mass index, and history of smoking, diabetes mellitus and cardiovascular disease to perform separate contrasts of non-hypertensive (normotensive or pre-hypertensive, N = 2332) and hypertensive (N = 2337) individuals and of normotensive (N = 741) and pre-hypertensive (N = 1581) individuals (p<0.05 after Bonferroni correction).ResultsThe brain white matter intracellular volume fraction was significantly lower, and isotropic volume fraction was higher in hypertensive relative to non-hypertensive individuals (N = 1559, each). The white matter isotropic volume fraction also was higher in pre-hypertensive than in normotensive individuals (N = 694, each) in the right superior longitudinal fasciculus and the right superior thalamic radiation, where the lower intracellular volume fraction was observed in the hypertensives relative to the non-hypertensive group.SignificancePathological processes associated with chronically elevated blood pressure are associated with imaging differences suggesting chronic alterations of white matter axonal structure that may affect cognitive functions even with pre-hypertension.

Journal article

Alfaro-Almagro F, Jenkinson M, Bangerter NK, Andersson JLR, Griffanti L, Douaud G, Sotiropoulos SN, Jbabdi S, Hernandez-Fernandez M, Vallee E, Vidaurre D, Webster M, McCarthy P, Rorden C, Daducci A, Alexander DC, Zhang H, Dragonu I, Matthews PM, Miller KL, Smith SMet al., 2017, Image processing and Quality Control for the first 10,000 brain imaging datasets from UK Biobank., NeuroImage, Vol: 166, Pages: 400-424, ISSN: 1053-8119

UK Biobank is a large-scale prospective epidemiological study with all data accessible to researchers worldwide. It is currently in the process of bringing back 100,000 of the original participants for brain, heart and body MRI, carotid ultrasound and low-dose bone/fat x-ray. The brain imaging component covers 6 modalities (T1, T2 FLAIR, susceptibility weighted MRI, Resting fMRI, Task fMRI and Diffusion MRI). Raw and processed data from the first 10,000 imaged subjects has recently been released for general research access. To help convert this data into useful summary information we have developed an automated processing and QC (Quality Control) pipeline that is available for use by other researchers. In this paper we describe the pipeline in detail, following a brief overview of UK Biobank brain imaging and the acquisition protocol. We also describe several quantitative investigations carried out as part of the development of both the imaging protocol and the processing pipeline.

Journal article

Matthews PM, 2017, Advanced MRI measures like DTI or fMRI should be outcome measures in future clinical trials - NO, Multiple Sclerosis Journal, Vol: 23, Pages: 1456-1458, ISSN: 1352-4585

Deep ultraviolet (DUV) light sources are used to neutralise isolated test masses in highly sensitive space-based gravitational experiments. An example is the LISA Pathfinder charge management system, which uses low-pressure mercury lamps. A future gravitational-wave observatory such as eLISA will use UV light-emitting diodes (UV LEDs), which offer numerous advantages over traditional discharge lamps. Such devices have limited space heritage but are now available from a number of commercial suppliers. Here we report on a test campaign that was carried out to quantify the general properties of three types of commercially available UV LEDs and demonstrate their suitability for use in space. Testing included general electrical and UV output power measurements, spectral stability, pulsed performance and temperature dependence, as well as thermal vacuum, radiation and vibration survivability.

Journal article

Peeters LM, Lamers I, Valkenborg D, Feys P, Somers V, Spooren A, Popescu V, Hens N, Matthews PM, Thalheim C, Van Wijmeersch B, Hellings Net al., 2017, Towards personalized therapy through extensive longitudinal follow-up using a multidisciplinary data infrastructure for people with MS: a-proof-of-concept study, 7th Joint European-Committee-for-Treatment-and-Research-in-Multiple-Sclerosis (ECTRIMS)-Americas-Committee-for-Treatment-and-Research-in-Multiple-Sclerosis (ACTRIMS), Publisher: SAGE PUBLICATIONS LTD, Pages: 934-935, ISSN: 1352-4585

Conference paper

Datta G, Colasanti A, Rabiner EA, Gunn RN, Malik O, Ciccarelli O, Nicholas R, Van Vlierberghe E, Van Hecke W, Searle G, Santos-Ribeiro A, Matthews PMet al., 2017, Neuroinflammation and its relationship to changes in brain volume and white matter lesions in multiple sclerosis, Brain, Vol: 140, Pages: 2927-2938, ISSN: 1460-2156

Brain magnetic resonance imaging is an important tool in the diagnosis and monitoring of multiple sclerosis patients. However, magnetic resonance imaging alone provides limited information for predicting an individual patient’s disability progression. In part, this is because magnetic resonance imaging lacks sensitivity and specificity for detecting chronic diffuse and multi-focal inflammation mediated by activated microglia/macrophages. The aim of this study was to test for an association between 18 kDa translocator protein brain positron emission tomography signal, which arises largely from microglial activation, and measures of subsequent disease progression in multiple sclerosis patients. Twenty-one patients with multiple sclerosis (seven with secondary progressive disease and 14 with a relapsing remitting disease course) underwent T1- and T2-weighted and magnetization transfer magnetic resonance imaging at baseline and after 1 year. Positron emission tomography scanning with the translocator protein radioligand 11C-PBR28 was performed at baseline. Brain tissue and lesion volumes were segmented from the T1- and T2-weighted magnetic resonance imaging and relative 11C-PBR28 uptake in the normal-appearing white matter was estimated as a distribution volume ratio with respect to a caudate pseudo-reference region. Normal-appearing white matter distribution volume ratio at baseline was correlated with enlarging T2-hyperintense lesion volumes over the subsequent year (ρ = 0.59, P = 0.01). A post hoc analysis showed that this association reflected behaviour in the subgroup of relapsing remitting patients (ρ = 0.74, P = 0.008). By contrast, in the subgroup of secondary progressive patients, microglial activation at baseline was correlated with later progression of brain atrophy (ρ = 0.86, P = 0.04). A regression model including the baseline normal-appearing white matter distribution volume ratio, T2 lesion volume and normal-appearing white matter magnet

Journal article

Coffey S, Lewandowski AJ, Garratt S, Meijer R, Lynum S, Bedi R, Paterson J, Yaqub M, Noble JA, Neubauer S, Petersen SE, Allen N, Sudlow C, Collins R, Matthews PM, Leeson Pet al., 2017, Protocol and quality assurance for carotid imaging in 100,000 participants of UK Biobank: development and assessment., European Journal of Preventive Cardiology, Vol: 24, Pages: 1799-1806, ISSN: 2047-4873

Background Ultrasound imaging is able to quantify carotid arterial wall structure for the assessment of cerebral and cardiovascular disease risks. We describe a protocol and quality assurance process to enable carotid imaging at large scale that has been developed for the UK Biobank Imaging Enhancement Study of 100,000 individuals. Design An imaging protocol was developed to allow measurement of carotid intima-media thickness from the far wall of both common carotid arteries. Six quality assurance criteria were defined and a web-based interface (Intelligent Ultrasound) was developed to facilitate rapid assessment of images against each criterion. Results and conclusions Excellent inter and intra-observer agreements were obtained for image quality evaluations on a test dataset from 100 individuals. The image quality criteria then were applied in the UK Biobank Imaging Enhancement Study. Data from 2560 participants were evaluated. Feedback of results to the imaging team led to improvement in quality assurance, with quality assurance failures falling from 16.2% in the first two-month period examined to 6.4% in the last. Eighty per cent had all carotid intima-media thickness images graded as of acceptable quality, with at least one image acceptable for 98% of participants. Carotid intima-media thickness measures showed expected associations with increasing age and gender. Carotid imaging can be performed consistently, with semi-automated quality assurance of all scans, in a limited timeframe within a large scale multimodality imaging assessment. Routine feedback of quality control metrics to operators can improve the quality of the data collection.

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

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

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

Suzuki H, Gao H, Bai W, Evangelou E, Glocker B, O'Regan DP, Elliott P, Matthews PMet al., 2017, Hypertension and white matter microstructures in healthy participants in UK Biobank, Publisher: OXFORD UNIV PRESS, Pages: 248-249, ISSN: 0195-668X

Conference paper

Dong H, Supratak A, Pan W, Wu C, Matthews PM, Guo Yet al., 2017, Mixed neural network approach for temporal sleep stage classification, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol: 26, Pages: 324-333, ISSN: 1534-4320

This paper proposes a practical approach to addressing limitations posed by using of single-channel electroencephalography (EEG) for sleep stage classification. EEG-based characterizations of sleep stage progression contribute the diagnosis and monitoring of the many pathologies of sleep. Several prior reports explored ways of automating the analysis of sleep EEG and of reducing the complexity of the data needed for reliable discrimination of sleep stages at lower cost in the home. However, these reports have involved recordings from electrodes placed on the cranial vertex or occiput, which are both uncomfortable and difficult to position. Previous studies of sleep stage scoring that used only frontal electrodes with a hierarchical decision tree motivated this paper, in which we have taken advantage of rectifier neural network for detecting hierarchical features and long short-term memory (LSTM) network for sequential data learning to optimize classification performance with single-channel recordings. After exploring alternative electrode placements, we found a comfortable configuration of a single-channel EEG on the forehead and have shown that it can be integrated with additional electrodes for simultaneous recording of the electrooculogram (EOG). Evaluation of data from 62 people (with 494 hours sleep) demonstrated better performance of our analytical algorithm than is available from existing approaches with vertex or occipital electrode placements. Use of this recording configuration with neural network deconvolution promises to make clinically indicated home sleep studies practical.

Journal article

Edison P, Mayers J, Calsolaro V, Fan Z, Turton S, Venkataraman A, Femminella G, Gunn R, Rabiner E, Matthews P, Tyacke R, Nutt Det al., 2017, Dementia Platform U.K. Experimental medicine: human in vivo astroglial activation in early Alzheimer’s disease, Alzheimer's and Dementia, Vol: 13, Pages: P1073-P1074, ISSN: 1552-5260

Journal article

Nie L, Yang X, Matthews PM, Xu ZW, Guo YKet al., 2017, Inferring functional connectivity in fMRI using minimum partial correlation, International Journal of Automation and Computing, Vol: 14, Pages: 371-385, ISSN: 1751-8520

Functional connectivity has emerged as a promising approach to study the functional organisation of the brain and to define features for prediction of brain state. The most widely used method for inferring functional connectivity is Pearson-s correlation, but it cannot differentiate direct and indirect effects. This disadvantage is often avoided by computing the partial correlation between two regions controlling all other regions, but this method suffers from Berkson-s paradox. Some advanced methods, such as regularised inverse covariance, have been applied. However, these methods usually depend on some parameters. Here we propose use of minimum partial correlation as a parameter-free measure for the skeleton of functional connectivity in functional magnetic resonance imaging (fMRI). The minimum partial correlation between two regions is the minimum of absolute values of partial correlations by controlling all possible subsets of other regions. Theoretically, there is a direct effect between two regions if and only if their minimum partial correlation is non-zero under faithfulness and Gaussian assumptions. The elastic PC-algorithm is designed to efficiently approximate minimum partial correlation within a computational time budget. The simulation study shows that the proposed method outperforms o thers in most cases and its application is illustrated using a resting-state fMRI dataset from the human connectome project.

Journal article

Owen DRJ, Narayan N, Wells L, Healy L, Smyth E, Rabiner EA, Galloway D, Williams JB, Lehr J, Mandhir H, Peferoen LAN, Taylor PC, Amor S, Antel JP, Matthews PM, Moore CSet al., 2017, Pro-inflammatory activation of primary microglia and macrophages increases 18kDa Translocator Protein (TSPO) expression in rodents but not humans, Journal of Cerebral Blood Flow and Metabolism, Vol: 37, Pages: 2679-2690, ISSN: 1559-7016

The 18kDa Translocator Protein (TSPO) is the most commonly used tissue-specific marker of inflammation in positron emission tomography (PET) studies. It is expressed in myeloid cells such as microglia and macrophages, and in rodent myeloid cells expression increases with cellular activation. We assessed the effect of myeloid cell activation on TSPO gene expression in both primary human and rodent microglia and macrophages in vitro, and also measured TSPO radioligand binding with 3H-PBR28 in primary human macrophages. As observed previously, we found that TSPO expression increases (∼9-fold) in rodent-derived macrophages and microglia upon pro-inflammatory stimulation. However, TSPO expression does not increase with classical pro-inflammatory activation in primary human microglia (fold change 0.85 [95% CI 0.58–1.12], p = 0.47). In contrast, pro-inflammatory activation of human monocyte-derived macrophages is associated with a reduction of both TSPO gene expression (fold change 0.60 [95% CI 0.45–0.74], p = 0.02) and TSPO binding site abundance (fold change 0.61 [95% CI 0.49–0.73], p < 0.0001). These findings have important implications for understanding the biology of TSPO in activated macrophages and microglia in humans. They are also clinically relevant for the interpretation of PET studies using TSPO targeting radioligands, as they suggest changes in TSPO expression may reflect microglial and macrophage density rather than activation phenotype.

Journal article

Wilman HR, Kelly M, Garratt S, Matthews PM, Milanesi M, Herlihy A, Gyngell M, Neubauer S, Bell JD, Banerjee R, Thomas ELet al., 2017, Correction: Characterisation of liver fat in the UK Biobank cohort, PLoS ONE, Vol: 12, Pages: e0176867-e0176867, ISSN: 1932-6203

[This corrects the article DOI: 10.1371/journal.pone.0172921.].

Journal article

Gafson A, Craner MJ, Matthews PM, 2017, Personalised medicine for multiple sclerosis care., Multiple Sclerosis Journal, Vol: 23, Pages: 362-369, ISSN: 1477-0970

Treatments with a range of efficacy and risk of adverse events have become available for the management of multiple sclerosis (MS). However, now the heterogeneity of clinical expression and responses to treatment pose major challenges to improving patient care. Selecting and managing the drug best balancing benefit and risk demands a new focus on the individual patient. Personalised medicine for MS is based on improving the precision of diagnosis for each patient in order to capture prognosis and provide an evidence-based framework for predicting treatment response and personalising patient monitoring. It involves development of predictive models involving the integration of clinical and biological data with an understanding of the impact of disease on the lives of individual patients. Here, we provide a brief, selective review of challenges to personalisation of the management of MS and suggest an agenda for stakeholder engagement and research to address them.

Journal article

Wilman HR, Kelly M, Garratt S, Matthews PM, Milanesi M, Herlihy A, Gyngell M, Neubauer S, Bell JD, Banerjee R, Thomas ELet al., 2017, Characterisation of liver fat in the UK Biobank cohort, PLOS ONE, Vol: 12, ISSN: 1932-6203

Non-alcoholicfattyliverdiseaseandtheriskof progressionto steatohepatitis,cirrhosisandhepatocellularcarcinomahavebeenidentifiedasmajorpublichealthconcerns.Wehavedemonstratedthefeasibilityandpotentialvalueof measuringliverfatcontentbymagneticresonanceimaging(MRI)in a largepopulationin thisstudyof 4,949participants(aged45–73years)in theUKBiobankimagingenhancement. Despiterequirementsforonlya single( 3min)scanof eachsubject,liverfatwasableto bemeasuredastheMRIprotondensityfatfraction(PDFF)withanoverallsuccessrateof 96.4%.Theoverallhepaticfatdistributionwascentredbetween1–2%,andwashighlyskewedtowardshigherfatcontent.ThemeanPDFFwas3.91%,andmedian2.11%.Analysisof PDFFin conjunctionwithotherdatafieldsavailablefromtheUKBiobankResourceshowedassociationsof increasedliverfatwithgreaterage,BMI,weightgain,highbloodpressureandType2 diabetes.SubjectswithBMIlessthan25kg/m2hada lowrisk(5%)of highliverfat(PDFF>5.5%),whereasin thehigherBMIpopulation(>30kg/m2) theprevalenceof highliverfatwasapproximately1 in 3. Thesedatasuggestthatpopulationscreeningto identifypeoplewithhighPDFFis possibleandcouldbecosteffective.MRIbasedPDFFis aneffectivemethodforthis.Finally,althoughcrosssectional,thisstudysuggeststheutilityof thePDFFmeasurement withinUKBiobank,particularlyforapplicationsto elucidatingriskfactorsthroughassociationswithprospec-tivelyacquireddataonclinicaloutcomesof liverdiseases,includingnon-alcoholicfattyliverdisease.

Journal article

Shenkin SD, Pernet C, Nichols TE, Poline J-B, Matthews PM, van der Lugt A, Mackay C, Lanyon L, Mazoyer B, Boardman JP, Thompson PM, Fox N, Marcus DS, Sheikh A, Cox SR, Anblagan D, Job DE, Dickie DA, Rodriguez D, Wardlaw JMet al., 2017, Improving data availability for brain image biobanking in healthy subjects: Practice-based suggestions from an international multidisciplinary working group, NEUROIMAGE, Vol: 153, Pages: 399-409, ISSN: 1053-8119

Journal article

Giovannoni G, Cutter G, Sormani MP, Belachew S, Hyde R, Koendgen H, Knappertz V, Tomic D, Leppert D, Herndon R, Wheeler-Kingshott CAM, Ciccarelli O, Selwood D, di Cantogno EV, Ben-Amor A-F, Matthews P, Carassiti D, Baker D, Schmierer Ket al., 2017, Is multiple sclerosis a length-dependent central axonopathy? The case for therapeutic lag and the asynchronous progressive MS hypotheses., Mult Scler Relat Disord, Vol: 12, Pages: 70-78

Trials of anti-inflammatory therapies in non-relapsing progressive multiple sclerosis (MS) have been stubbornly negative except recently for an anti-CD20 therapy in primary progressive MS and a S1P modulator siponimod in secondary progressive MS. We argue that this might be because trials have been too short and have focused on assessing neuronal pathways, with insufficient reserve capacity, as the core component of the primary outcome. Delayed neuroaxonal degeneration primed by prior inflammation is not expected to respond to disease-modifying therapies targeting MS-specific mechanisms. However, anti-inflammatory therapies may modify these damaged pathways, but with a therapeutic lag that may take years to manifest. Based on these observations we propose that clinically apparent neurodegenerative components of progressive MS may occur in a length-dependent manner and asynchronously. If this hypothesis is confirmed it may have major implications for the future design of progressive MS trials.

Journal article

Kalk NJ, Guo Q, Owen D, Cherian R, Erritzoe D, Gilmour A, Ribeiro AS, McGonigle J, Waldman A, Matthews P, Cavanagh J, McInnes I, Dar K, Gunn R, Rabiner EA, Lingford-Hughes ARet al., 2017, Decreased hippocampal translocator protein (18 kDa) expression in alcohol dependence: a [(11)C]PBR28 PET study, Translational Psychiatry, Vol: 7, ISSN: 2158-3188

Repeated withdrawal from alcohol is clinically associated with progressive cognitive impairment. Microglial activation occurring during pre-clinical models of alcohol withdrawal is associated with learning deficits. We investigated whether there was microglial activation in recently detoxified alcohol-dependent patients (ADP), using [(11)C]PBR28 positron emission tomography (PET), selective for the 18kDa translocator protein (TSPO) highly expressed in activated microglia and astrocytes. We investigated the relationship between microglial activation and cognitive performance. Twenty healthy control (HC) subjects (45±13; M:F 14:6) and nine ADP (45±6, M:F 9:0) were evaluated. Dynamic PET data were acquired for 90 min following an injection of 331±15 MBq [(11)C]PBR28. Regional volumes of distribution (VT) for regions of interest (ROIs) identified a priori were estimated using a two-tissue compartmental model with metabolite-corrected arterial plasma input function. ADP had an ~20% lower [(11)C]PBR28 VT, in the hippocampus (F(1,24) 5.694; P=0.025), but no difference in VT in other ROIs. Hippocampal [(11)C]PBR28 VT was positively correlated with verbal memory performance in a combined group of HC and ADP (r=0.720, P<0.001), an effect seen in HC alone (r=0.738; P=0.001) but not in ADP. We did not find evidence for increased microglial activation in ADP, as seen pre-clinically. Instead, our findings suggest lower glial density or an altered activation state with lower TSPO expression. The correlation between verbal memory and [(11)C]PBR28 VT, raises the possibility that abnormalities of glial function may contribute to cognitive impairment in ADP.

Journal article

Poldrack RA, Baker CI, Durnez J, Gorgolewski KJ, Matthews PM, Munafo MR, Nichols TE, Poline J-B, Vul E, Yarkoni Tet al., 2017, Scanning the horizon: towards transparent and reproducible neuroimaging research, Nature Reviews Neuroscience, Vol: 18, Pages: 115-126, ISSN: 1471-0048

Functional neuroimaging techniques have transformed our ability to probe the neurobiological basis of behaviour and are increasingly being applied by the wider neuroscience community. However, concerns have recently been raised that the conclusions that are drawn from some human neuroimaging studies are either spurious or not generalizable. Problems such as low statistical power, flexibility in data analysis, software errors and a lack of direct replication apply to many fields, but perhaps particularly to functional MRI. Here, we discuss these problems, outline current and suggested best practices, and describe how we think the field should evolve to produce the most meaningful and reliable answers to neuroscientific questions.

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=00501273&limit=30&person=true&page=2&respub-action=search.html