Higher order spectra based support vector machine for arrhythmia classification

Heart rate variability (HRV) refers to the regulation of the sinoatrial node, the natural pacemaker of the heart by the sympathetic and parasympathetic branches of the autonomic nervous system. HRV analysis is an important tool to observe the heart’s ability to respond to normal regulatory impulses that affect its rhythm. Like many bio-signals, HRV signals are non-linear in nature. Higher order spectral analysis (HOS) is known to be a good tool for the analysis of non-linear systems and provides good noise immunity. A computer-based arrhythmia detection system of cardiac states is very useful in diagnostics and disease management. In this work, we studied the identification of the HRV signals using features derived from HOS. These features were fed to the support vector machine (SVM) for classification. Our proposed system can classify the normal and other four classes of arrhythmia with an average accuracy of more than 85%.

A model to estimate the durability performance of both normal and light-weight concrete

There has been an increasing focus on the development of test methods to evaluate the durability performance of concrete. This paper contributes to this focus by presenting a study that evaluates the effect of water accessible porosity and oven-dry unit weight on the resistance of both normal and light-weight concrete to chloride-ion penetration. Based on the experimental results and regression analyses, empirical models are established to correlate the total charge passed and the chloride migration coefficient with the basic properties of concrete such as water accessible porosity, oven dry unit weight, and compressive strength. These equations can be broadly applied to both normal and lightweight aggregate concretes. The model was also validated by an independent set of experimental results from two different concrete mixtures. The model provides a very good estimate on the concrete’s durability performance in respect to the resistance to chloride ion penetration.

Classification of echolocation calls from 14 species of bat by Support Vector Machines and Ensembles of Neural Networks

Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%) consistently outperformed SVMs (mean identification rate – 87%). Correct classification rates produced by the ENNs varied from 91% to 100%; calls from six species were correctly identified with 100% accuracy. Calls from the five species of Myotis, a genus whose species are considered difficult to distinguish acoustically, had correct identification rates that varied from 91 – 100%. Five parameters were most important for classifying calls correctly while seven others contributed little to classification performance.

Estimating the burden of disease attributable to excess body weight in South Africa in 2000

Objective. To estimate the burden of disease attributable to excess body weight using the body mass index (BMI), by age and sex, in South Africa in 2000. Design. World Health Organization comparative risk assessment (CRA) methodology was followed. Re-analysis of the 1998 South Africa Demographic and Health Survey data provided mean BMI estimates by age and sex. Populationattributable fractions were calculated and applied to revised burden of disease estimates. Monte Carlo simulation-modelling techniques were used for the uncertainty analysis. Setting. South Africa. Subjects. Adults 30 years of age. Outcome measures. Deaths and disability-adjusted life years (DALYs) from ischaemic heart disease, ischaemic stroke, hypertensive disease, osteoarthritis, type 2 diabetes mellitus, and selected cancers. Results. Overall, 87% of type 2 diabetes, 68% of hypertensive disease, 61% of endometrial cancer, 45% of ischaemic stroke, 38% of ischaemic heart disease, 31% of kidney cancer, 24% of osteoarthritis, 17% of colon cancer, and 13% of postmenopausal breast cancer were attributable to a BMI 21 kg/m2. Excess body weight is estimated to have caused 36 504 deaths (95% uncertainty interval 31 018 - 38 637) or 7% (95% uncertainty interval 6.0 - 7.4%) of all deaths in 2000, and 462 338 DALYs (95% uncertainty interval 396 512 - 478 847) or 2.9% of all DALYs (95% uncertainty interval 2.4 - 3.0%). The burden in females was approximately double that in males. Conclusions. This study shows the importance of recognising excess body weight as a major risk to health, particularly among females, highlighting the need to develop, implement and evaluate comprehensive interventions to achieve lasting change in the determinants and impact of excess body weight.

Reliable Fault Diagnosis for Low-Speed Bearings Using Individually Trained Support Vector Machines With Kernel Discriminative Feature Analysis

This paper proposes a highly reliable fault diagnosis approach for low-speed bearings. The proposed approach first extracts wavelet-based fault features that represent diverse symptoms of multiple low-speed bearing defects. The most useful fault features for diagnosis are then selected by utilizing a genetic algorithm (GA)-based kernel discriminative feature analysis cooperating with one-against-all multicategory support vector machines (OAA MCSVMs). Finally, each support vector machine is individually trained with its own feature vector that includes the most discriminative fault features, offering the highest classification performance. In this study, the effectiveness of the proposed GA-based kernel discriminative feature analysis and the classification ability of individually trained OAA MCSVMs are addressed in terms of average classification accuracy. In addition, the proposedGA- based kernel discriminative feature analysis is compared with four other state-of-the-art feature analysis approaches. Experimental results indicate that the proposed approach is superior to other feature analysis methodologies, yielding an average classification accuracy of 98.06% and 94.49% under rotational speeds of 50 revolutions-per-minute (RPM) and 80 RPM, respectively. Furthermore, the individually trained MCSVMs with their own optimal fault features based on the proposed GA-based kernel discriminative feature analysis outperform the standard OAA MCSVMs, showing an average accuracy of 98.66% and 95.01% for bearings under rotational speeds of 50 RPM and 80 RPM, respectively.

Error and variance bounds on sigmoidal neurons with weight and input errors

Bounds on the expectation and variance of errors at the output of a multilayer feedforward neural network with perturbed weights and inputs are derived. It is assumed that errors in weights and inputs to the network are statistically independent and small. The bounds obtained are applicable to both digital and analogue network implementations and are shown to be of practical value.

An analysis of DNA methylation in human adipose tissue reveals differential modification of obesity genes before and after gastric bypass and weight loss

Background Environmental factors can influence obesity by epigenetic mechanisms. Adipose tissue plays a key role in obesity-related metabolic dysfunction, and gastric bypass provides a model to investigate obesity and weight loss in humans. Results Here, we investigate DNA methylation in adipose tissue from obese women before and after gastric bypass and significant weight loss. In total, 485,577 CpG sites were profiled in matched, before and after weight loss, subcutaneous and omental adipose tissue. A paired analysis revealed significant differential methylation in omental and subcutaneous adipose tissue. A greater proportion of CpGs are hypermethylated before weight loss and increased methylation is observed in the 3′ untranslated region and gene bodies relative to promoter regions. Differential methylation is found within genes associated with obesity, epigenetic regulation and development, such as CETP, FOXP2, HDAC4, DNMT3B, KCNQ1 and HOX clusters. We identify robust correlations between changes in methylation and clinical trait, including associations between fasting glucose and HDAC4, SLC37A3 and DENND1C in subcutaneous adipose. Genes investigated with differential promoter methylation all show significantly different levels of mRNA before and after gastric bypass. Conclusions This is the first study reporting global DNA methylation profiling of adipose tissue before and after gastric bypass and associated weight loss. It provides a strong basis for future work and offers additional evidence for the role of DNA methylation of adipose tissue in obesity.

Weight and weight change following breast cancer : evidence from a prospective, population-based, breast cancer cohort study

Background: While weight gain following breast cancer is considered common, results supporting these findings are dated. This work describes changes in body weight following breast cancer over 72 months, compares weight with normative data and explores whether weight changes over time are associated with personal, diagnostic, treatment or behavioral characteristics. Methods: A population-based sample of 287 Australian women diagnosed with early-stage invasive breast cancer was assessed prospectively at six, 12, 18 and 72 months post-surgery. Weight was clinically measured and linear mixed models were used to explore associations between weight and participant characteristics (collected via self-administered questionnaire). Those with BMI changes of one or more units were considered to have experienced clinically significant changes in weight. Results: More than half (57%) of participants were overweight or obese at 6 months post-surgery, and by 72 months post-surgery 68% of women were overweight or obese. Among those who gained more weight than age-matched norms, clinically significant weight gain between 6 and 18 months and 6 and 72 months post-surgery was observed in 24% and 39% of participants, respectively (median [range] weight gain: 3.9kg [2.0-11.3kg] and 5.2kg [0.6-28.7], respectively). Clinically-significant weight losses were observed in up to 24% of the sample (median [range] weight loss between 6 and 72 months post-surgery: -6.4kg [-1.9--24.6kg]). More extensive lymph node removal, being treated on the non-dominant side, receiving radiation therapy and lower physical activity levels at 6 months was associated with higher body weights post-breast cancer (group differences >3kg; all p<0.05). Conclusions: While average weight gain among breast cancer survivors in the long-term is small, subgroups of women experience greater gains linked with adverse health and above that experienced by age-matched counterparts. Weight change post-breast cancer is a contemporary public health issue and the integration of healthy weight education and support into standard breast cancer care has potential to significantly improve the length and quality of cancer survivorship.

Food neophobia and its association with diet quality and weight in children aged 24 months : a cross sectional study

Background Food neophobia, the rejection of unknown or novel foods, may result in poor dietary patterns. This study investigates the cross-sectional relationship between neophobia in children aged 24 months and variety of fruit and vegetable consumption, intake of discretionary foods and weight. Methods Secondary analysis of data from 330 parents of children enrolled in the NOURISH RCT (control group only) and SAIDI studies was performed using data collected at child age 24 months. Neophobia was measured at 24 months using the Child Food Neophobia Scale (CFNS). The cross-sectional associations between total CFNS score and fruit and vegetable variety, discretionary food intake and BMI (Body Mass Index) Z-score were examined via multiple regression models; adjusting for significant covariates. Results At 24 months, more neophobic children were found to have lower variety of fruits (β=-0.16, p=0.003) and vegetables (β=-0.29, p<0.001) but have a greater proportion of daily energy from discretionary foods (β=0.11, p=0.04). There was no significant association between BMI Z-score and CFNS score. Conclusions Neophobia is associated with poorer dietary quality. Results highlight the need for interventions to (1) begin early to expose children to a wide variety of nutritious foods before neophobia peaks and (2) enable health professionals to educate parents on strategies to overcome neophobia.

DS-VBF : dual sink vector-based routing protocol for underwater wireless sensor network

Underwater wireless sensor networks (UWSNs) have become the seat of researchers' attention recently due to their proficiency to explore underwater areas and design different applications for marine discovery and oceanic surveillance. One of the main objectives of each deployed underwater network is discovering the optimized path over sensor nodes to transmit the monitored data to onshore station. The process of transmitting data consumes energy of each node, while energy is limited in UWSNs. So energy efficiency is a challenge in underwater wireless sensor network. Dual sinks vector based forwarding (DS-VBF) takes both residual energy and location information into consideration as priority factors to discover an optimized routing path to save energy in underwater networks. The modified routing protocol employs dual sinks on the water surface which improves network lifetime. According to deployment of dual sinks, packet delivery ratio and the average end to end delay are enhanced. Based on our simulation results in comparison with VBF, average end to end delay reduced more than 80%, remaining energy increased 10%, and the increment of packet reception ratio was about 70%.

What determines weight loss behaviour and body weight satisfaction in Australia? Evidence from the HILDA survey

This thesis undertakes an empirical investigation to identify factors that influence the decision to undertake weight loss behaviour using the nationally representative HILDA dataset. Although many factors influenced the decision, the findings suggested that body weight satisfaction was the greatest determinant of weight loss dieting. This thesis therefore conducted a further empirical study to analyse the determinants of body weight satisfaction. A rank-hypothesis was found to better predict variation in body weight satisfaction levels than the absolute value of the individual's Body Mass Index (BMI) or the relative-norm hypothesis, which are commonly reported in the literature.