Improving Near-Miss Event Detection Rate at Railway Level Crossings

Even though crashes between trains and road users are rare events at railway level crossings, they are one of the major safety concerns for the Australian railway industry. Nearmiss events at level crossings occur more frequently, and can provide more information about factors leading to level crossing incidents. In this paper we introduce a video analytic approach for automatically detecting and localizing vehicles from cameras mounted on trains for detecting near-miss events. To detect and localize vehicles at level crossings we extract patches from an image and classify each patch for detecting vehicles. We developed a region proposals algorithm for generating patches, and we use a Convolutional Neural Network (CNN) for classifying each patch. To localize vehicles in images we combine the patches that are classified as vehicles according to their CNN scores and positions. We compared our system with the Deformable Part Models (DPM) and Regions with CNN features (R-CNN) object detectors. Experimental results on a railway dataset show that the recall rate of our proposed system is 29% higher than what can be achieved with DPM or R-CNN detectors.

Experimental investigation of wear modes of hardened rail material used in Australian heavy-haul railways

This work investigates the effects of contact pressure and geometry in rolling-contact wear tests by using discs with different radii of curvature to simulate the varying contact conditions that may be typically found in the field. The tests were conducted without any significant amount of traction, but micro slip was still observed due to contact deformation. Moreover, variation of contact pressure was observed due to contact patch elongation and diameter reduction. Rolling contact fatigue, adhesive and sliding wear were observed on the curved contact interface. The development of different wear regimes and material removal phenomena were analyzed using microscopic images in order to broaden the understanding of the wear mechanisms occurring in the rail-wheel contact.

Schizophrenia Genetic Variants are not Associated with Intelligence

Background Schizophrenia is associated with lower pre-morbid intelligence (IQ) in addition to (pre-morbid) cognitive decline. Both schizophrenia and IQ are highly heritable traits. Therefore, we hypothesized that genetic variants associated with schizophrenia, including copy number variants (CNVs) and a polygenic schizophrenia (risk) score (PSS), may influence intelligence. Method IQ was estimated with the Wechsler Adult Intelligence Scale (WAIS). CNVs were determined from single nucleotide polymorphism (SNP) data using the QuantiSNP and PennCNV algorithms. For the PSS, odds ratios for genome-wide SNP data were calculated in a sample collected by the Psychiatric Genome-Wide Association Study (GWAS) Consortium (8690 schizophrenia patients and 11 831 controls). These were used to calculate individual PSSs in our independent sample of 350 schizophrenia patients and 322 healthy controls. Results Although significantly more genes were disrupted by deletions in schizophrenia patients compared to controls (p = 0.009), there was no effect of CNV measures on IQ. The PSS was associated with disease status (R 2 = 0.055, p = 2.1 × 10 -7) and with IQ in the entire sample (R 2 = 0.018, p = 0.0008) but the effect on IQ disappeared after correction for disease status. Conclusions Our data suggest that rare and common schizophrenia-associated variants do not explain the variation in IQ in healthy subjects or in schizophrenia patients. Thus, reductions in IQ in schizophrenia patients may be secondary to other processes related to schizophrenia risk. © Cambridge University Press 2013.

Dichotomy of food and inhalant allergen sensitization in eosinophilic esophagitis

Background: Eosinophilic esophagitis (EE) is an emerging condition where patients commonly present with symptoms of gastroesophageal reflux disease and fail to respond adequately to anti-reflux therapy. Food allergy is currently recognized as the main immunological cause of EE; recent evidence suggests an etiological role for inhalant allergens. The presence of EE appears to be associated with other atopic illnesses. Objectives: To report the sensitization profile of both food and inhalant allergens in our EE patient cohort in relation to age, and to profile the prevalence of other allergic conditions in patients with EE. Method: The study prospectively analyzed allergen sensitization profiles using skin prick tests to common food allergens and inhalant allergens in 45 children with EE. Patch testing to common food allergens was performed on 33 patients in the same cohort. Comorbidity of atopic eczema, asthma, allergic rhinitis and anaphylaxis were obtained from patient history. Results: Younger patients with EE showed more IgE and patch sensitization to foods while older patients showed greater IgE sensitization to inhalant allergens. The prevalence of atopic eczema, allergic rhinitis and asthma was significantly increased in our EE cohort compared with the general Australian population. A total of 24% of our cohort of patients with EE had a history of anaphylaxis. Conclusion: In children with EE, the sensitization to inhalant allergens increases with age, particularly after 4 years. Also, specific enquiry about severe food reactions in patients presenting with EE is strongly recommended as it appears this patient group has a high incidence of anaphylaxis. © 2007 The Authors.

Repeatable Condition-Invariant Visual Odometry for Sequence-Based Place Recognition

This paper describes a vision-only system for place recognition in environments that are tra- versed at different times of day, when chang- ing conditions drastically affect visual appear- ance, and at different speeds, where places aren’t visited at a consistent linear rate. The ma- jor contribution is the removal of wheel-based odometry from the previously presented algo- rithm (SMART), allowing the technique to op- erate on any camera-based device; in our case a mobile phone. While we show that the di- rect application of visual odometry to our night- time datasets does not achieve a level of perfor- mance typically needed, the VO requirements of SMART are orthogonal to typical usage: firstly only the magnitude of the velocity is required, and secondly the calculated velocity signal only needs to be repeatable in any one part of the environment over day and night cycles, but not necessarily globally consistent. Our results show that the smoothing effect of motion constraints is highly beneficial for achieving a locally consis- tent, lighting-independent velocity estimate. We also show that the advantage of our patch-based technique used previously for frame recogni- tion, surprisingly, does not transfer to VO, where SIFT demonstrates equally good performance. Nevertheless, we present the SMART system us- ing only vision, which performs sequence-base place recognition in extreme low-light condi- tions where standard 6-DOF VO fails and that improves place recognition performance over odometry-less benchmarks, approaching that of wheel odometry.

Steel–concrete–steel sandwich system in Arctic offshore structure: Materials, experiments, and design

The gravity based structure (GBS) with external Steel–Concrete–Steel (SCS) sandwich ice-resistant wall has been developed for the Arctic oil and gas drilling. This paper firstly reported the experimental studies on the mechanical properties of steel and concretes under Arctic low temperature. With the test data, design equations were developed to incorporate the influences of the low temperature on these mechanical properties. Two types of Arctic GBS structure with flower-conical SCS sandwich shell type and plate type of ice-resistant wall have been developed for the Arctic offshore structure. Besides the studies on the materials, two SCS sandwich prototype shells and plates were, respectively, prepared and tested under patch loading that simulated the localized ice-contact pressure. The structural behaviors of the SCS sandwich structure under patch loading were reported and discussions were made on the influences of different parameters on the structural behavior of the structure. Analytical models were developed to predict the punching shear resistances of the SCS sandwich structure through modifying the code provisions. The accuracies of the developed analytical models were checked through validations against 27 tests in the literature. Corresponding design procedures on resistances of SCS sandwich structure were recommended based on these discussions and validations.

Learning with symmetric label noise: The importance of being unhinged

Convex potential minimisation is the de facto approach to binary classification. However, Long and Servedio [2008] proved that under symmetric label noise (SLN), minimisation of any convex potential over a linear function class can result in classification performance equivalent to random guessing. This ostensibly shows that convex losses are not SLN-robust. In this paper, we propose a convex, classification-calibrated loss and prove that it is SLN-robust. The loss avoids the Long and Servedio [2008] result by virtue of being negatively unbounded. The loss is a modification of the hinge loss, where one does not clamp at zero; hence, we call it the unhinged loss. We show that the optimal unhinged solution is equivalent to that of a strongly regularised SVM, and is the limiting solution for any convex potential; this implies that strong l2 regularisation makes most standard learners SLN-robust. Experiments confirm the unhinged loss’ SLN-robustness.

The role of 25-hydroxyvitamin D deficiency in promoting insulin resistance and inflammation in patients with chronic kidney disease: A randomised controlled trial

BACKGROUND Approximately 50% of patients with stage 3 Chronic Kidney Disease are 25-hydroxyvitamin D insufficient, and this prevalence increases with falling glomerular filtration rate. Vitamin D is now recognised as having pleiotropic roles beyond bone and mineral homeostasis, with the vitamin D receptor and metabolising machinery identified in multiple tissues. Worryingly, recent observational data has highlighted an association between hypovitaminosis D and increased cardiovascular mortality, possibly mediated via vitamin D effects on insulin resistance and inflammation. The main hypothesis of this study is that oral Vitamin D supplementation will ameliorate insulin resistance in patients with Chronic Kidney Disease stage 3 when compared to placebo. Secondary hypotheses will test whether this is associated with decreased inflammation and bone/adipocyte-endocrine dysregulation. METHODS/DESIGN This study is a single-centre, double-blinded, randomised, placebo-controlled trial. Inclusion criteria include; estimated glomerular filtration rate 30-59 ml/min/1.73 m(2); aged >or=18 on entry to study; and serum 25-hydroxyvitamin D levels <75 nmol/L. Patients will be randomised 1:1 to receive either oral cholecalciferol 2000IU/day or placebo for 6 months. The primary outcome will be an improvement in insulin sensitivity, measured by hyperinsulinaemic euglycaemic clamp. Secondary outcome measures will include serum parathyroid hormone, cytokines (Interleukin-1beta, Interleukin-6, Tumour Necrosis Factor alpha), adiponectin (total and High Molecular Weight), osteocalcin (carboxylated and under-carboxylated), peripheral blood mononuclear cell Nuclear Factor Kappa-B p65 binding activity, brachial artery reactivity, aortic pulse wave velocity and waveform analysis, and indirect calorimetry. All outcome measures will be performed at baseline and end of study. DISCUSSION To date, no randomised controlled trial has been performed in pre-dialysis CKD patients to study the correlation between vitamin D status with supplementation, insulin resistance and markers of adverse cardiovascular risk. We remain hopeful that cholecalciferol may be a safe intervention, with health benefits beyond those related to bone-mineral homeostasis. TRIAL REGISTRATION Australian and New Zealand Clinical Trials Registry ACTRN12609000246280.