He T.,Royal Brompton Hospital |
He T.,Imperial College London |
Zhang J.,Defence College of Management and Technology |
Carpenter J.-P.,Royal Brompton Hospital |
And 8 more authors.
Journal of Magnetic Resonance Imaging | Year: 2013
Purpose: To propose an automated truncation method for myocardial T2* measurement and evaluate this method on a large population of patients with iron loading in the heart and scanned at multiple magnetic resonance imaging (MRI) centers. Materials and Methods: A total of 550 thalassemia patients were scanned at 20 international centers using a variety of MR scanners (Siemens, Philips, or GE). A single mid-ventricular short axis slice was imaged. All patient data were anonymized before the T2* were measured by expert observers using standard techniques. These same datasets were then retrospectively processed using the proposed automated truncation method by another independent observer and the resulting T2* measurements were compared with those of expert readings. Results: The T2* measurements using the automated method showed good agreement with those measured by expert observers using standard techniques (P = 0.95) with a low coefficient of variation (1.6%). Conclusion: This study demonstrates feasibility and good reproducibility of a new automated truncation method for myocardial T2* measurement. This approach simplifies the overall analysis and can be easily incorporated into T2* analysis software to facilitate further development of a fully automated myocardial tissue iron quantification. © 2012 Wiley Periodicals, Inc.
Maphanga R.R.,University of Limpopo |
Sayle D.C.,Defence College of Management and Technology |
Sayle T.X.T.,Defence College of Management and Technology |
Ngoepe P.E.,University of Limpopo
Physical Chemistry Chemical Physics | Year: 2011
Various polymorphs of MnO2 are widely used as electrode materials in Li/MnO2 batteries. Electrolytic manganese dioxide (EMD) is the most electrochemically active form of MnO2 and is very difficult to characterize. Their structural details are still largely unknown owing to the poor quality of X-ray diffraction (XRD) patterns obtained from most MnO2 samples. Simulated amorphisation and crystallization technique was used to derive microstructural models for Li-MnO2 which included most microstructural details that one would expect to find in the real material. Specifically, pyrolusite-MnO2, comprising about 25000 atoms, was amorphised (strain-induced) under molecular dynamics (MD) and different concentrations of lithium ions were inserted. Each system was then crystallized under MD simulation. The resulting models conformed to the pyrolusite polymorph, with microstructural features including: extensive micro-twinning and more general grain-boundaries, stacking faults, dislocations and isolated point defects and defect clusters. Molecular graphical images, showing the atom positions for the microstructural features together with simulated XRD patterns they give rise to, are presented and compared with measured XRD. The calculated XRD are in accord with experiment thus validating the structural models. © 2011 the Owner Societies.
Wu J.,Defence College of Management and Technology |
Wu J.,Imperial College London |
Barahona M.,Imperial College London |
Tan Y.-J.,Defence College of Management and Technology |
Deng H.-Z.,Defence College of Management and Technology
International Journal of Systems Science | Year: 2011
It has been recently proposed that natural connectivity can be used to efficiently characterise the robustness of complex networks. The natural connectivity quantifies the redundancy of alternative routes in the network by evaluating the weighted number of closed walks of all lengths and can be seen as an average eigenvalue obtained from the graph spectrum. In this article, we explore both analytically and numerically the natural connectivity of regular ring lattices and regular random graphs obtained through degree-preserving random rewirings from regular ring lattices. We reformulate the natural connectivity of regular ring lattices in terms of generalised Bessel functions and show that the natural connectivity of regular ring lattices is independent of network size and increases with K monotonically. We also show that random regular graphs have lower natural connectivity, and are thus less robust, than regular ring lattices. © 2011 Taylor & Francis.
Shanmugavel M.,Defence College of Management and Technology |
Tsourdos A.,Defence College of Management and Technology |
White B.A.,Defence College of Management and Technology
2010 15th International Conference on Methods and Models in Automation and Robotics, MMAR 2010 | Year: 2010
This paper discusses path planning of a Micro Air Vehicle (UAV) with collision avoidance in a three dimensional space. The 3D space is chosen as an urban environment where the buildings are sparsely located. The process is divided into two phases: (i) Route planning where the buildings are isolated to form a route, which is followed by (ii) Path planning in which simple Dubins paths are employed to follow through the waypoints generated by the route planner. The first phase creates a network of waypoints in between the buildings which meets the minimum distance required for safe manoeuvre. Delaunay triangulation is used to isolate the buildings from each other and to produce safe waypoints between them. Once the waypoints are assured safe to pass through, the path planner uses Dubins path to produce the flyable and safe paths. ©2010 IEEE.
Tan S.,Defence College of Management and Technology |
Xia Q.,Defence College of Management and Technology |
Basu A.,University of Alberta |
Lou J.,Defence College of Management and Technology |
Zhang M.,Defence College of Management and Technology
Journal of Modern Optics | Year: 2014
We propose a new spatial mapping method for hybrid vision systems consisting of an omnidirectional camera and an active perspective camera. We calculate the relative position between the optical centers of the two cameras based on pre-acquisition parameters of two sample points from the observed scene, avoiding the difficulty and error in direct measurement, and at the same time using fewer points. We derive the formula of the relative position between the omnidirectional cameras optical center and the active cameras optical center from the distance between two points, and give a method for calculating the corresponding pan and tilt angles of the active camera from the coordinates in the omnidirectional image. Simulation and real experiments show that the two-point spatial mapping method has higher estimation accuracy than the homography calibration method, and at the same time has higher consistency among test points. © 2014 Taylor & Francis.
Wenman M.R.,Imperial College London |
Chard-Tuckey P.R.,Defence College of Management and Technology
International Journal of Plasticity | Year: 2010
The use of 3D digital image correlation (DIC) has been used to capture the Lüders strains in a low carbon ferritic steel. Results were used to calibrate and compare with finite element (FE) results based on a constitutive plasticity model, capable of yield drop behaviour and therefore Lüders strains, by Zhang et al. (2001). Tensile tests were carried out at several strain rates to characterise the material behaviour. The results of these tests were used to fit parameters in the constitutive plasticity model. The FE model was then tested on a complex loading situation of in-plane compression of a compact tension (CT) specimen. The FE model predicts the shape and formation of the Lüders bands well. This FE model, using Zhang's constitutive plasticity model, was used to predict the residual stress profile to compare with standard elastic-plastic isotropic hardening models with no yield point. The yield point reduced both the predicted peak tensile stress, at the notch root, and the amount of plastic strain. In regions where the plastic strain was of a similar size to the Lüders strain the stress profiles were perturbed from flat profiles predicted by the standard elastic-plastic hardening models.
Wang R.,University of Sheffield |
Wang R.,Defence College of Management and Technology |
Purshouse R.C.,University of Sheffield |
Fleming P.J.,University of Sheffield
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013
Various multi-objective evolutionary algorithms (MOEAs) have been developed to help a decision maker (DM) search for his/her preferred solutions to multi-objective problems. However, none of these approaches has catered simultaneously for the two fundamental ways that DM can specify his/her preferences: weights and aspiration levels. In this paper, we propose an approach named iPICEA-g that allows the DM to specify his preference in either format. iPICEA-g is based on the preference-inspired co-evolutionary algorithm (PICEA-g). Solutions are guided toward regions of interest (ROIs) to the DM by co-evolving sets of goal vectors exclusively generated in the ROIs. Moreover, a friendly decision making technique is developed for interaction with the optimization process: the DM specifies his preferences easily by interactively brushing his preferred regions in the objective space. No direct elicitation of numbers is required, reducing the cognitive burden on DM. The performance of iPICEA-g is tested on a set of benchmark problems and is shown to be good. © 2013 Springer-Verlag.
Speight L.R.,Defence College of Management and Technology |
Rowland D.,Defence College of Management and Technology
Military Operations Research | Year: 2010
This paper marks the completion of a process that started with the construction of a model of the approach phase of an infantry assault, based on the evidence of peacetime trials. In the next phaseitmovedontoinclude the behavioural effects induced by the combat environment, as established by historical analysis. In this final phase a representation of group morale is added, and the simulation is extended in order to cover the final assault and the chances of attack success. The paper first gives an account of the few available formal studies of these last two topics. It then moves on to review some of the most important features that can be noted in the very many anecdotal accounts of combat that have been published. A description is then provided of the several components of the developed model. Itis shown that key outputs from this modelarewellinlinewiththecorresponding quantified results of historical analysis. Illustrations are given of some of these outputs, and simple equations are provided that will produce predictions of such criteria as casualties, or chances of attack success, as functions of opposing numbers, terrain or force effectiveness. The paper concludes with a discussion of some wider implications that follow from this paper, including the difficulties of extrapolating from one military culture to another.
Tsitiridis A.,Defence College of Management and Technology |
Richardson M.,Defence College of Management and Technology |
Yuen P.W.T.,Defence College of Management and Technology
Engineering Intelligent Systems | Year: 2012
This preliminary study concentrates on the refinement of early models in cortex-like machine vision, in particular Feature Hierarchy Library (FHLib), in order to remedy some of its drawbacks and introduces a salient feature approach to generic object recognition. Enhancements in three areas, a) extraction of features from the most salient region of interest (ROI) and their rearrangement in a ranked manner, rather than random extraction over the whole image, b) exploitation of larger patches to improve spatial resolutions and performance, c) a more versatile template matching technique, have been the main contributions of the present work. Under different classification methods, the improved model is validated using 4 different types of datasets and shows better recognition accuracy over the original FHLib model. This improvement has remained consistent even after the reduction of the library of features by approximately 40%. ©2012 CRL Publishing Ltd.
Bengtsson L.,Karolinska Institutet |
Bengtsson L.,Flowminder Foundation |
Gaudart J.,Aix - Marseille University |
Lu X.,Karolinska Institutet |
And 8 more authors.
Scientific Reports | Year: 2015
Effective response to infectious disease epidemics requires focused control measures in areas predicted to be at high risk of new outbreaks. We aimed to test whether mobile operator data could predict the early spatial evolution of the 2010 Haiti cholera epidemic. Daily case data were analysed for 78 study areas from October 16 to December 16, 2010. Movements of 2.9 million anonymous mobile phone SIM cards were used to create a national mobility network. Two gravity models of population mobility were implemented for comparison. Both were optimized based on the complete retrospective epidemic data, available only after the end of the epidemic spread. Risk of an area experiencing an outbreak within seven days showed strong dose-response relationship with the mobile phone-based infectious pressure estimates. The mobile phone-based model performed better (AUC 0.79) than the retrospectively optimized gravity models (AUC 0.66 and 0.74, respectively). Infectious pressure at outbreak onset was significantly correlated with reported cholera cases during the first ten days of the epidemic (p < 0.05). Mobile operator data is a highly promising data source for improving preparedness and response efforts during cholera outbreaks. Findings may be particularly important for containment efforts of emerging infectious diseases, including high-mortality influenza strains. © 2015, Nature Publishing Group. All rights reserved.