Ranking Spatial and Temporal Hotspots for Road Safety Interventions Using a Nonlinear Aggregation Approach

In recent years, the concept of a composite performance index, brought from economic and business statistics, has gained popularity in the field of road safety. The construction of the Composite Safety Performance Index (CSPI) involves the following key steps: the selection of the most appropriate indicators to be aggregated and the method used to aggregate them.

Over the last decade, various aggregation methods for estimating the CSPI have been suggested in the literature. However, recent studies indicates that most of these methods suffer from many deficiencies at both the theoretical and operational level; these include the correlation and compensability between indicators, as well as their high “degree of freedom” which enables one to readily manipulate them to produce desired outcomes.

The purpose of this study is to introduce an alternative aggregation method for the estimation of the CSPI, which is free from the aforementioned deficiencies. In contrast with the current aggregation methods, which generally use linear combinations of road safety indicators to estimate a CSPI, the approach advocated in this study is based on non-linear combinations of indicators and can be summarized into the following two main steps: the pairwise comparison of road safety indicators and the development of marginal and composite road safety performance functions. The introduced method has been successfully applied to identify and rank temporal and spatial hotspots for Northern Ireland, using road traffic collision data recorded in the UK STATs19 database. The obtained results highlight the promising features of the proposed approach including its stability and consistency, which enables significantly reduced deficiencies associated with the current aggregation methods. Progressively, the introduced method could evolve into an intelligent support system for road safety assessment.

BIIL 284 reduces neutrophil numbers but increases P. aeruginosa bacteremia and inflammation in mouse lungs

BACKGROUND: A clinical study to investigate the leukotriene B(4) (LTB(4))-receptor antagonist BIIL 284 in cystic fibrosis (CF) patients was prematurely terminated due to a significantly increased risk of adverse pulmonary events. We aimed to establish the effect of BIIL284 in models of Pseudomonas aeruginosa lung infection, thereby contributing to a better understanding of what could have led to adverse pulmonary events in CF patients.

METHODS: P. aeruginosa DNA in the blood of CF patients during and after acute pulmonary exacerbations and in stable patients with non-CF bronchiectasis (NCFB) and healthy individuals was assessed by PCR. The effect of BIIL 284 treatment was tested in an agar bead murine model of P. aeruginosa lung infection. Bacterial count and inflammation were evaluated in lung and other organs.

RESULTS: Most CF patients (98%) and all patients with NCFB and healthy individuals had negative P. aeruginosa DNA in their blood. Similarly, the P. aeruginosa-infected mice showed bacterial counts in the lung but not in the blood or spleen. BIIL 284 treatment decreased pulmonary neutrophils and increased P. aeruginosa numbers in mouse lungs leading to significantly higher bacteremia rates and lung inflammation compared to placebo treated animals.

CONCLUSIONS: Decreased airway neutrophils induced lung proliferation and severe bacteremia in a murine model of P. aeruginosa lung infection. These data suggest that caution should be taken when administering anti-inflammatory compounds to patients with bacterial infections.

Quantum Chromatic Numbers via Operator Systems

We define several new types of quantum chromatic numbers of a graph and characterize them in terms of operator system tensor products. We establish inequalities between these chromatic numbers and other parameters of graphs studied in the literature and exhibit a link between them and non-signalling correlation boxes.

Iso-Quality of Service: Fairly Ranking Servers for Real-Time Data Analytics

We present a mathematically rigorous Quality-of-Service (QoS) metric which relates the achievable quality of service metric (QoS) for a real-time analytics service to the server energy cost of offering the service. Using a new iso-QoS evaluation methodology, we scale server resources to meet QoS targets and directly rank the servers in terms of their energy-efficiency and by extension cost of ownership. Our metric and method are platform-independent and enable fair comparison of datacenter compute servers with significant architectural diversity, including micro-servers. We deploy our metric and methodology to compare three servers running financial option pricing workloads on real-life market data. We find that server ranking is sensitive to data inputs and desired QoS level and that although scale-out micro-servers can be up to two times more energy-efficient than conventional heavyweight servers for the same target QoS, they are still six times less energy efficient than high-performance computational accelerators.

An Efficient LS-SVM-Based Method for Fuzzy System Construction

This paper proposes an efficient learning mechanism to build fuzzy rule-based systems through the construction of sparse least-squares support vector machines (LS-SVMs). In addition to the significantly reduced computational complexity in model training, the resultant LS-SVM-based fuzzy system is sparser while offers satisfactory generalization capability over unseen data. It is well known that the LS-SVMs have their computational advantage over conventional SVMs in the model training process; however, the model sparseness is lost, which is the main drawback of LS-SVMs. This is an open problem for the LS-SVMs. To tackle the nonsparseness issue, a new regression alternative to the Lagrangian solution for the LS-SVM is first presented. A novel efficient learning mechanism is then proposed in this paper to extract a sparse set of support vectors for generating fuzzy IF-THEN rules. This novel mechanism works in a stepwise subset selection manner, including a forward expansion phase and a backward exclusion phase in each selection step. The implementation of the algorithm is computationally very efficient due to the introduction of a few key techniques to avoid the matrix inverse operations to accelerate the training process. The computational efficiency is also confirmed by detailed computational complexity analysis. As a result, the proposed approach is not only able to achieve the sparseness of the resultant LS-SVM-based fuzzy systems but significantly reduces the amount of computational effort in model training as well. Three experimental examples are presented to demonstrate the effectiveness and efficiency of the proposed learning mechanism and the sparseness of the obtained LS-SVM-based fuzzy systems, in comparison with other SVM-based learning techniques.

Diabetes in the young - a global view and worldwide estimates of numbers of children with type 1 diabetes

This paper describes the methodology, results and limitations of the 2013 International Diabetes Federation (IDF) Atlas (6th edition) estimates of the worldwide numbers of prevalent cases of type 1 diabetes in children (<15 years). The majority of relevant information in the published literature is in the form of incidence rates derived from registers of newly diagnosed cases. Studies were graded on quality criteria and, if no information was available in the published literature, extrapolation was used to assign a country the rate from an adjacent country with similar characteristics. Prevalence rates were then derived from these incidence rates and applied to United Nations 2012 Revision population estimates for 2013 for each country to obtain estimates of the number of prevalent cases. Data availability was highest for the countries in Europe (76%) and lowest for the countries in sub-Saharan Africa (8%). The prevalence estimates indicate that there are almost 500,000 children aged under 15 years with type 1 diabetes worldwide, the largest numbers being in Europe (129,000) and North America (108,700). Countries with the highest estimated numbers of new cases annually were the United States (13,000), India (10,900) and Brazil (5000). Compared with the prevalence estimates made in previous editions of the IDF Diabetes Atlas, the numbers have increased in most of the IDF Regions, often reflecting the incidence rate increases that have been well-documented in many countries. Monogenic diabetes is increasingly being recognised among those with clinical features of type 1 or type 2 diabetes as genetic studies become available, but population-based data on incidence and prevalence show wide variation due to lack of standardisation in the studies. Similarly, studies on type 2 diabetes in childhood suggest increased incidence and prevalence in many countries, especially in Indigenous peoples and ethnic minorities, but detailed population-based studies remain limited.

Identification of Dynamics of Surface Suction Over an Airfoil at Low Reynolds Numbers

This paper deals with identification of dynamics in suction control of airfoils for low Reynolds number regimes (8 x 10^4 - 5 x 10^5). In particular, the dynamics of interest is the map that relates suction pressure and surface pressure. Identification of such dynamics is of use to a variety of active control applications including suction control in small/medium wind turbines which operate in these Reynolds number regimes. Prior research has largely focused on higher Reynolds number regimes, creating a need for such a study. Towards identifying the said dynamic relations, experiments were conducted on NACA0012 airfoil in a wind tunnel. The dynamic relation between suction and surface pressure was identified as an overdamped second order system.

Privacy-Protected Facial Biometric Verification via Fuzzy Forest Learning

Although visual surveillance has emerged as an effective technolody for public security, privacy has become an issue of great concern in the transmission and distribution of surveillance videos. For example, personal facial images should not be browsed without permission. To cope with this issue, face image scrambling has emerged as a simple solution for privacyrelated applications. Consequently, online facial biometric verification needs to be carried out in the scrambled domain thus bringing a new challenge to face classification. In this paper, we investigate face verification issues in the scrambled domain and propose a novel scheme to handle this challenge. In our proposed method, to make feature extraction from scrambled face images robust, a biased random subspace sampling scheme is applied to construct fuzzy decision trees from randomly selected features, and fuzzy forest decision using fuzzy memberships is then obtained from combining all fuzzy tree decisions. In our experiment, we first estimated the optimal parameters for the construction of the random forest, and then applied the optimized model to the benchmark tests using three publically available face datasets. The experimental results validated that our proposed scheme can robustly cope with the challenging tests in the scrambled domain, and achieved an improved accuracy over all tests, making our method a promising candidate for the emerging privacy-related facial biometric applications.