Speech recognition in adverse environments using lip information

The performance of automatic speech recognition systems deteriorates in the presence of noise. One known solution is to incorporate video information with an existing acoustic speech recognition system. We investigate the performance of the individual acoustic and visual sub-systems and then examine different ways in which the integration of the two systems may be performed. The system is to be implemented in real time on a Texas Instruments' TMS320C80 DSP.

The use of speech and lip modalities for robust speaker verification under adverse conditions

Investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. We have previously shown (Int. Conf. on Acoustics, Speech and Signal Proc., vol. 6, pp. 3693-3696, May 1998) that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms either subsystem individually. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

Parametric differences between a real-world distributed denial-of-service attack and a flash event

Distributed Denial-of-Service (DDoS) attacks continue to be one of the most pernicious threats to the delivery of services over the Internet. Not only are DDoS attacks present in many guises, they are also continuously evolving as new vulnerabilities are exploited. Hence accurate detection of these attacks still remains a challenging problem and a necessity for ensuring high-end network security. An intrinsic challenge in addressing this problem is to effectively distinguish these Denial-of-Service attacks from similar looking Flash Events (FEs) created by legitimate clients. A considerable overlap between the general characteristics of FEs and DDoS attacks makes it difficult to precisely separate these two classes of Internet activity. In this paper we propose parameters which can be used to explicitly distinguish FEs from DDoS attacks and analyse two real-world publicly available datasets to validate our proposal. Our analysis shows that even though FEs appear very similar to DDoS attacks, there are several subtle dissimilarities which can be exploited to separate these two classes of events.

Dynamic texture reconstruction from sparse codes for unusual event detection in crowded scenes

Unusual event detection in crowded scenes remains challenging because of the diversity of events and noise. In this paper, we present a novel approach for unusual event detection via sparse reconstruction of dynamic textures over an overcomplete basis set, with the dynamic texture described by local binary patterns from three orthogonal planes (LBPTOP). The overcomplete basis set is learnt from the training data where only the normal items observed. In the detection process, given a new observation, we compute the sparse coefficients using the Dantzig Selector algorithm which was proposed in the literature of compressed sensing. Then the reconstruction errors are computed, based on which we detect the abnormal items. Our application can be used to detect both local and global abnormal events. We evaluate our algorithm on UCSD Abnormality Datasets for local anomaly detection, which is shown to outperform current state-of-the-art approaches, and we also get promising results for rapid escape detection using the PETS2009 dataset.

The role of idea novelty and relatedness in nascent ventures

The study of venture idea characteristics and the contextual fit between venture ideas and individuals are key research goals in entrepreneurship (Davidsson, 2004). However, to date there has been limited scholarly attention given to these phenomena. Accordingly, this study aims to help fill the gap by investigating the importance of novelty and relatedness of venture ideas in entrepreneurial firms. On the premise that new venture creation is a process and that research should be focused on the early stages of the venturing process, this study primarily focuses its attention on examining how venture idea novelty and relatedness affect the performance in the venture creation process. Different types and degrees of novelty are considered here. Relatedness is shown to be based on individuals’ prior knowledge and resource endowment. Performance in the venture creation process is evaluated according to four possible outcomes: making progress, getting operational, being terminated and achieving positive cash flow. A theoretical model is developed demonstrating the relationship between these variables along with the investment of time and money. Several hypotheses are developed to be tested. Among them, it is hypothesised that novelty hinders short term performance in the venture creation process. On the other hand knowledge and resource relatedness are hypothesised to promote performance. An experimental study was required in order to understand how different types and degrees of novelty and relatedness of venture ideas affect the attractiveness of venture ideas in the eyes of experienced entrepreneurs. Thus, the empirical work in this thesis was based on two separate studies. In the first one, a conjoint analysis experiment was conducted on 32 experienced entrepreneurs in order to ascertain attitudinal preferences regarding venture idea attractiveness based on novelty, relatedness and potential financial gains. This helped to estimate utility values for different levels of different attributes of venture ideas and their relative importance in the attractiveness. The second study was a longitudinal investigation of how venture idea novelty and relatedness affect the performance in the venture creation process. The data for this study is from the Comprehensive Australian Study for Entrepreneurial Emergence (CAUSEE) project that has been established in order to explore the new venture creation process in Australia. CAUSEE collects data from a representative sample of over 30,000 households in Australia using random digit dialling (RDD) telephone interviews. From these cases, data was collected at two points in time during a 12 month period from 493 firms, who are currently involved in the start-up process. Hypotheses were tested and inferences were derived through descriptive statistics, confirmatory factor analysis and structural equation modelling. Results of study 1 indicate that venture idea characteristics have a role in the attractiveness and entrepreneurs prefer to introduce a moderate degree of novelty across all types of venture ideas concerned. Knowledge relatedness is demonstrated to be a more significant factor in attractiveness than resource relatedness. Results of study 2 show that the novelty hinders nascent venture performance. On the other hand, resource relatedness has a positive impact on performance unlike knowledge relatedness which has none. The results of these studies have important implications for potential entrepreneurs, investors, researchers, consultants etc. by developing a better understanding in the venture creation process and its success factors in terms of both theory and practice.

Unusual Event Detection in Crowded Scenes Using Bag of LBPs in spatio-temporal patches

Modelling events in densely crowded environments remains challenging, due to the diversity of events and the noise in the scene. We propose a novel approach for anomalous event detection in crowded scenes using dynamic textures described by the Local Binary Patterns from Three Orthogonal Planes (LBP-TOP) descriptor. The scene is divided into spatio-temporal patches where LBP-TOP based dynamic textures are extracted. We apply hierarchical Bayesian models to detect the patches containing unusual events. Our method is an unsupervised approach, and it does not rely on object tracking or background subtraction. We show that our approach outperforms existing state of the art algorithms for anomalous event detection in UCSD dataset.

Planning volunteer responses to low-volume mass gatherings: do event characteristics predict patient workload?

INTRODUCTION: Workforce planning for first aid and medical coverage of mass gatherings is hampered by limited research. In particular, the characteristics and likely presentation patterns of low-volume mass gatherings of between several hundred to several thousand people are poorly described in the existing literature. OBJECTIVES: This study was conducted to: 1. Describe key patient and event characteristics of medical presentations at a series of mass gatherings, including events smaller than those previously described in the literature; 2. Determine whether event type and event size affect the mean number of patients presenting for treatment per event, and specifically, whether the 1:2,000 deployment rule used by St John Ambulance Australia is appropriate; and 3. Identify factors that are predictive of injury at mass gatherings. METHODS: A retrospective, observational, case-series design was used to examine all cases treated by two Divisions of St John Ambulance (Queensland) in the greater metropolitan Brisbane region over a three-year period (01 January 2002-31 December 2004). Data were obtained from routinely collected patient treatment forms completed by St John officers at the time of treatment. Event-related data (e.g., weather, event size) were obtained from event forms designed for this study. Outcome measures include: total and average number of patient presentations for each event; event type; and event size category. Descriptive analyses were conducted using chi-square tests, and mean presentations per event and event type were investigated using Kruskal-Wallis tests. Logistic regression analyses were used to identify variables independently associated with injury presentation (compared with non-injury presentations). RESULTS: Over the three-year study period, St John Ambulance officers treated 705 patients over 156 separate events. The mean number of patients who presented with any medical condition at small events (less than or equal to 2,000 attendees) did not differ significantly from that of large (>2,000 attendees) events (4.44 vs. 4.67, F = 0.72, df = 1, 154, p = 0.79). Logistic regression analyses indicated that presentation with an injury compared with non-injury was independently associated with male gender, winter season, and sporting events, even after adjusting for relevant variables. CONCLUSIONS: In this study of low-volume mass gatherings, a similar number of patients sought medical treatment at small (<2,000 patrons) and large (>2,000 patrons) events. This demonstrates that for low-volume mass gatherings, planning based solely on anticipated event size may be flawed, and could lead to inappropriate levels of first-aid coverage. This study also highlights the importance of considering other factors, such as event type and patient characteristics, when determining appropriate first-aid resourcing for low-volume events. Additionally, identification of factors predictive of injury presentations at mass gatherings has the potential to significantly enhance the ability of event coordinators to plan effective prevention strategies and response capability for these events.

The influence of ambient temperature on birth outcomes in Brisbane, Australia

Lately, there has been increasing interest in the association between temperature and adverse birth outcomes including preterm birth (PTB) and stillbirth. PTB is a major predictor of many diseases later in life, and stillbirth is a devastating event for parents and families. The aim of this study was to assess the seasonal pattern of adverse birth outcomes, and to examine possible associations of maternal exposure to temperature with PTB and stillbirth. We also aimed to identify if there were any periods of the pregnancy where exposure to temperature was particularly harmful. A retrospective cohort study design was used and we retrieved individual birth records from the Queensland Health Perinatal Data Collection Unit for all singleton births (excluding twins and triplets) delivered in Brisbane between 1 July 2005 and 30 June 2009. We obtained weather data (including hourly relative humidity, minimum and maximum temperature) and air-pollution data (including PM10, SO2 and O3) from the Queensland Department of Environment and Resource Management. We used survival analyses with the time-dependent variables of temperature, humidity and air pollution, and the competing risks of stillbirth and live birth. To assess the monthly pattern of the birth outcomes, we fitted month of pregnancy as a time-dependent variable. We examined the seasonal pattern of the birth outcomes and the relationship between exposure to high or low temperatures and birth outcomes over the four lag weeks before birth. We further stratified by categorisation of PTB: extreme PTB (< 28 weeks of gestation), PTB (28–36 weeks of gestation), and term birth (≥ 37 weeks of gestation). Lastly, we examined the effect of temperature variation in each week of the pregnancy on birth outcomes. There was a bimodal seasonal pattern in gestation length. After adjusting for temperature, the seasonal pattern changed from bimodal, to only one peak in winter. The risk of stillbirth was statistically significant lower in March compared with January. After adjusting for temperature, the March trough was still statistically significant and there was a peak in risk (not statistically significant) in winter. There was an acute effect of temperature on gestational age and stillbirth with a shortened gestation for increasing temperature from 15 °C to 25 °C over the last four weeks before birth. For stillbirth, we found an increasing risk with increasing temperatures from 12 °C to approximately 20 °C, and no change in risk at temperatures above 20 °C. Certain periods of the pregnancy were more vulnerable to temperature variation. The risk of PTB (28–36 weeks of gestation) increased as temperatures increased above 21 °C. For stillbirth, the fetus was most vulnerable at less than 28 weeks of gestation, but there were also effects in 28–36 weeks of gestation. For fetuses of more than 37 weeks of gestation, increasing temperatures did not increase the risk of stillbirth. We did not find any adverse affects of cold temperature on birth outcomes in this cohort. My findings contribute to knowledge of the relationship between temperature and birth outcomes. In the context of climate change, this is particularly important. The results may have implications for public health policy and planning, as they indicate that pregnant women would decrease their risk of adverse birth outcomes by avoiding exposure to high temperatures and seeking cool environments during hot days.

Event-related brain potentials dissociate the developmental time-course of automatic numerical magnitude analysis and cognitive control functions during the first three years of primary school

In this study we set out to dissociate the developmental time course of automatic symbolic number processing and cognitive control functions in grade 1-3 British primary school children. Event-related potential (ERP) and behavioral data were collected in a physical size discrimination numerical Stroop task. Task-irrelevant numerical information was processed automatically already in grade 1. Weakening interference and strengthening facilitation indicated the parallel development of general cognitive control and automatic number processing. Relationships among ERP and behavioral effects suggest that control functions play a larger role in younger children and that automaticity of number processing increases from grade 1 to 3.

Microbial colonisation of human follicular fluid and adverse in vitro fertilisation outcomes

This study, investigating 263 women undergoing trans-vaginal oocyte retrieval for in vitro fertilisation (IVF) found that microorganisms colonising follicular fluid contributed to adverse IVF (pre-implantation) and pregnancy (post-implantation) outcomes including poor quality embryos, failed pregnancy and early pregnancy loss (< 37 weeks gestation). Some microorganisms also showed in vitro growth patterns in liquid media that appeared to be enhanced by the hormonal stimulation protocol used for oocyte retrieval. Elaborated cytokines within follicular fluid were also associated with adverse IVF outcomes. This study is imperative because infertility affects 16% of the human population and the numbers of couples needing assistance continues to increase. Despite significant improvements in the technical aspects of assisted reproductive technologies (ART), the live birth rate has not increased proportionally. Overt genital tract infection has been associated with both infertility and adverse pregnancy outcomes (including miscarriage and preterm birth) as a direct result of the infection or the host response to it. Importantly, once inflammation had become established, medical treatment often failed to prevent these significant adverse outcomes. Current evaluations of fertility focus on the ovary as a site of steroid hormone production and ovulation. However, infertility as a result of subclinical colonisation of the ovary has not been reported. Furthermore, identification of the microorganisms present in follicular fluid and the local cytokine profile may provide clinicians with an early indication of the prognosis for IVF treatment in infertile couples, thus allowing antimicrobial treatment and/or counselling about possible IVF failure. During an IVF cycle, multiple oocytes undergo maturation in vivo in response to hormonal hyperstimulation. Oocytes for in vitro insemination are collected trans-vaginally. The follicular fluid that bathes the maturing oocyte in vivo, usually is discarded as part of the IVF procedure, but provides a unique opportunity to investigate microbial causes of adverse IVF outcomes. Some previous studies have identified follicular fluid markers that predict IVF pregnancy outcomes. However, there have not been any detailed microbiological studies of follicular fluid. For this current study, paired follicular fluid and vaginal secretion samples were collected from women undergoing IVF cycles to determine whether microorganisms in follicular fluid were associated with adverse IVF outcomes. Microorganisms in follicular fluid were regarded as either "colonisers" or "contaminants"; colonisers, if they were unique to the follicular fluid sample, and contaminants if the same microorganisms were detected in the vaginal and follicular fluid samples indicating that the follicular fluid was merely contaminated during the oocyte retrieval process. Quite unexpectedly, by these criteria, we found that follicular fluid from approximately 30% of all subjects was colonised with bacteria. Fertile and infertile women with colonised follicular fluid had decreased embryo transfer rates and decreased pregnancy rates compared to women with contaminated follicular fluids. The observation that follicular fluid was not always sterile, but contained a diverse range of microorganisms, is novel. Many of the microorganisms we detected in follicular fluid are known opportunistic pathogens that have been detected in upper genital tract infections and are associated with adverse pregnancy outcomes. Bacteria were able to survive for at least 28 weeks in vitro, in cultures of follicular fluid. Within 10 days of establishing these in vitro cultures, several species (Lactobacillus spp., Bifidobacterium spp., Propionibacterium spp., Streptococcus spp. and Salmonella entericus) had formed biofilms. Biofilms play a major role in microbial pathogenicity and persistence. The propensity of microbial species to form biofilms in follicular fluid suggests that successful treatment of these infections with antimicrobials may be difficult. Bifidobacterium spp. grew, in liquid media, only if concentrations of oestradiol and progesterone were similar to those achieved in vivo during an IVF cycle. In contrast, the growth of Streptococcus agalactiae and Escherichia coli was inhibited or abolished by the addition of these hormones to culture medium. These data suggest that the likelihood of microorganisms colonising follicular fluid and the species of bacteria involved is influenced by the stage of the menstrual cycle and, in the case of IVF, the nature and dose of steroid hormones administered for the maturation of multiple oocytes in vivo. Our findings indicate that the elevated levels of steroid hormones during an IVF cycle may influence the microbial growth within follicular fluid, suggesting that the treatment itself will impact on the microflora present in the female upper genital tract during pre-conception and early post-conception phases of the cycle. The effect of the host immune response on colonising bacteria and on the outcomes of IVF also was investigated. White blood cells reportedly compose between 5% and 15% of the cell population in follicular fluid. The follicular membrane is semi-permeable and cells are actively recruited as part of the normal menstrual cycle and in response to microorganisms. A previous study investigated follicular fluid cytokines from infertile women and fertile oocyte donors undergoing IVF, and concluded that there were no significant differences in the cytokine concentrations between the two groups. However, other studies have reported differences in the follicular fluid cytokine levels associated with infertile women with endometriosis or polycystic ovary syndrome. In this study, elevated levels of interleukin (IL)-1 á, IL-1 â and vascular endothelial growth factor (VEGF) in vaginal fluid were associated with successful fertilisation, which may be useful marker for successful fertilisation outcomes for women trying to conceive naturally or prior to oocyte retrieval for IVF. Elevated levels of IL-6, IL-12p40, granulocyte colony stimulating factor (GCSF) and interferon-gamma (IFN ã) in follicular fluid were associated with successful embryo transfer. Elevated levels of pro-inflammatory IL-18 and decreased levels of anti-inflammatory IL-10 were identified in follicular fluid from women with idiopathic infertility. Successful fertilisation and implantation is dependent on a controlled pro-inflammatory environment, involving active recruitment of pro-inflammatory mediators to the genital tract as part of the menstrual cycle and early pregnancy. However, ongoing pregnancy requires an enhanced anti-inflammatory environment to ensure that the maternal immune system does not reject the semi-allergenic foetus. The pro-inflammatory skew in the follicular fluid of women with idiopathic infertility, correlates with normal rates of fertilisation, embryo discard and embryo transfer, observed for this cohort, which were similar to the outcomes observed for fertile women. However, their pregnancy rate was reduced compared to fertile women. An altered local immune response in follicular fluid may provide a means of explaining infertility in this cohort, previously defined as 'idiopathic'. This study has found that microorganisms colonising follicular fluid may have contributed to adverse IVF and pregnancy outcomes. Follicular fluid bathes the cumulus oocyte complex during the in vivo maturation process, and microorganisms in the fluid, their metabolic products or the local immune response to these microorganisms may result in damage to the oocytes, degradation of the cumulus or contamination of the IVF culture system. Previous studies that have discounted bacterial contamination of follicular fluid as a cause of adverse IVF outcomes failed to distinguish between bacteria that were introduced into the follicular fluid at the time of trans-vaginal oocyte retrieval and those that colonised the follicular fluid. Those bacteria that had colonised the fluid may have had time to form biofilms and to elicit a local immune response. Failure to draw this distinction has previously prevented consideration of bacterial colonisation of follicular fluid as a cause of adverse IVF outcomes. Several observations arising from this study are of significance to IVF programs. Follicular fluid is not always sterile and colonisation of follicular fluid is a cause of adverse IVF and pregnancy outcomes. Hormonal stimulation associated with IVF may influence whether follicular fluid is colonised and enhance the growth of specific species of bacteria within follicular fluid. Bacteria in follicular fluid may form biofilms and literature has reported that this may influence their susceptibility to antibiotics. Monitoring the levels of selected cytokines within vaginal secretions may inform fertilisation outcomes. This study has identified novel factors contributing to adverse IVF outcomes and that are most likely to affect also natural conception outcomes. Early intervention, possibly using antimicrobial or immunological therapies may reduce the need for ART and improve reproductive health outcomes for all women.

Application of Geographic Modeling Techniques to Quantify Spatial Access to Health Services Before and After an Acute Cardiac Event: The Cardiac ARIA Project

Background: Access to cardiac services is essential for appropriate implementation of evidence-based therapies to improve outcomes. The Cardiac Accessibility and Remoteness Index for Australia (Cardiac ARIA) aimed to derive an objective, geographic measure reflecting access to cardiac services. Methods: An expert panel defined an evidence-based clinical pathway. Using Geographic Information Systems (GIS), a numeric/alpha index was developed at two points along the continuum of care. The acute category (numeric) measured the time from the emergency call to arrival at an appropriate medical facility via road ambulance. The aftercare category (alpha) measured access to four basic services (family doctor, pharmacy, cardiac rehabilitation, and pathology services) when a patient returned to their community. Results: The numeric index ranged from 1 (access to principle referral center with cardiac catheterization service ≤ 1 hour) to 8 (no ambulance service, > 3 hours to medical facility, air transport required). The alphabetic index ranged from A (all 4 services available within 1 hour drive-time) to E (no services available within 1 hour). 13.9 million (71%) Australians resided within Cardiac ARIA 1A locations (hospital with cardiac catheterization laboratory and all aftercare within 1 hour). Those outside Cardiac 1A were over-represented by people aged over 65 years (32%) and Indigenous people (60%). Conclusion: The Cardiac ARIA index demonstrated substantial inequity in access to cardiac services in Australia. This methodology can be used to inform cardiology health service planning and the methodology could be applied to other common disease states within other regions of the world.

Mapping Services to Support a Patient's Journey through Evidence-Based Care Pathways After a Cardiac Event

Background: There are inequalities in geographical access and delivery of health care services in Australia, particularly for cardiovascular disease (CVD), Australia's major cause of death. Analyses and models that can inform and positively influence strategies to augment services and preventative measures are needed. The Cardiac-ARIA project is using geographical spatial technology (GIS) to develop a national index for each of Australia's 13,000 population centres. The index will describe the spatial distribution of CVD health care services available to support populations at risk, in a timely manner, after a major cardiac event. Methods: In the initial phase of the project, an expert panel of cardiologists and an emergency physician have identified key elements of national and international guidelines for management of acute coronary syndromes, cardiac arrest, life-threatening arrhythmias and acute heart failure, from the time of onset (potentially dial 000) to return from the hospital to the community (cardiac rehabilitation). Results: A systematic search has been undertaken to identify the geographical location of, and type of, cardiac services currently available. This has enabled derivation of a master dataset of necessary services, e.g. telephone networks, ambulance, RFDS, helicopter retrieval services, road networks, hospitals, general practitioners, medical community centres, pathology services, CCUs, catheterisation laboratories, cardio-thoracic surgery units and cardiac rehabilitation services. Conclusion: This unique and innovative project has the potential to deliver a powerful tool to both highlight and combat the burden of disease of CVD in urban and regional Australia.

A 2-h diagnostic protocol to assess patients with chest pain symptoms in the Asia-Pacific region (ASPECT): a prospective observational validation study

Background: Patients with chest pain contribute substantially to emergency department attendances, lengthy hospital stay, and inpatient admissions. A reliable, reproducible, and fast process to identify patients presenting with chest pain who have a low short-term risk of a major adverse cardiac event is needed to facilitate early discharge. We aimed to prospectively validate the safety of a predefined 2-h accelerated diagnostic protocol (ADP) to assess patients presenting to the emergency department with chest pain symptoms suggestive of acute coronary syndrome. Methods: This observational study was undertaken in 14 emergency departments in nine countries in the Asia-Pacific region, in patients aged 18 years and older with at least 5 min of chest pain. The ADP included use of a structured pre-test probability scoring method (Thrombolysis in Myocardial Infarction [TIMI] score), electrocardiograph, and point-of-care biomarker panel of troponin, creatine kinase MB, and myoglobin. The primary endpoint was major adverse cardiac events within 30 days after initial presentation (including initial hospital attendance). This trial is registered with the Australia-New Zealand Clinical Trials Registry, number ACTRN12609000283279. Findings: 3582 consecutive patients were recruited and completed 30-day follow-up. 421 (11•8%) patients had a major adverse cardiac event. The ADP classified 352 (9•8%) patients as low risk and potentially suitable for early discharge. A major adverse cardiac event occurred in three (0•9%) of these patients, giving the ADP a sensitivity of 99•3% (95% CI 97•9–99•8), a negative predictive value of 99•1% (97•3–99•8), and a specificity of 11•0% (10•0–12•2). Interpretation: This novel ADP identifies patients at very low risk of a short-term major adverse cardiac event who might be suitable for early discharge. Such an approach could be used to decrease the overall observation periods and admissions for chest pain. The components needed for the implementation of this strategy are widely available. The ADP has the potential to affect health-service delivery worldwide.