Application of data mining techniques in the prediction of coronary artery disease : use of anaesthesia time-series and patient risk factor data

The high morbidity and mortality associated with atherosclerotic coronary vascular disease (CVD) and its complications are being lessened by the increased knowledge of risk factors, effective preventative measures and proven therapeutic interventions. However, significant CVD morbidity remains and sudden cardiac death continues to be a presenting feature for some subsequently diagnosed with CVD. Coronary vascular disease is also the leading cause of anaesthesia related complications. Stress electrocardiography/exercise testing is predictive of 10 year risk of CVD events and the cardiovascular variables used to score this test are monitored peri-operatively. Similar physiological time-series datasets are being subjected to data mining methods for the prediction of medical diagnoses and outcomes. This study aims to find predictors of CVD using anaesthesia time-series data and patient risk factor data. Several pre-processing and predictive data mining methods are applied to this data. Physiological time-series data related to anaesthetic procedures are subjected to pre-processing methods for removal of outliers, calculation of moving averages as well as data summarisation and data abstraction methods. Feature selection methods of both wrapper and filter types are applied to derived physiological time-series variable sets alone and to the same variables combined with risk factor variables. The ability of these methods to identify subsets of highly correlated but non-redundant variables is assessed. The major dataset is derived from the entire anaesthesia population and subsets of this population are considered to be at increased anaesthesia risk based on their need for more intensive monitoring (invasive haemodynamic monitoring and additional ECG leads). Because of the unbalanced class distribution in the data, majority class under-sampling and Kappa statistic together with misclassification rate and area under the ROC curve (AUC) are used for evaluation of models generated using different prediction algorithms. The performance based on models derived from feature reduced datasets reveal the filter method, Cfs subset evaluation, to be most consistently effective although Consistency derived subsets tended to slightly increased accuracy but markedly increased complexity. The use of misclassification rate (MR) for model performance evaluation is influenced by class distribution. This could be eliminated by consideration of the AUC or Kappa statistic as well by evaluation of subsets with under-sampled majority class. The noise and outlier removal pre-processing methods produced models with MR ranging from 10.69 to 12.62 with the lowest value being for data from which both outliers and noise were removed (MR 10.69). For the raw time-series dataset, MR is 12.34. Feature selection results in reduction in MR to 9.8 to 10.16 with time segmented summary data (dataset F) MR being 9.8 and raw time-series summary data (dataset A) being 9.92. However, for all time-series only based datasets, the complexity is high. For most pre-processing methods, Cfs could identify a subset of correlated and non-redundant variables from the time-series alone datasets but models derived from these subsets are of one leaf only. MR values are consistent with class distribution in the subset folds evaluated in the n-cross validation method. For models based on Cfs selected time-series derived and risk factor (RF) variables, the MR ranges from 8.83 to 10.36 with dataset RF_A (raw time-series data and RF) being 8.85 and dataset RF_F (time segmented time-series variables and RF) being 9.09. The models based on counts of outliers and counts of data points outside normal range (Dataset RF_E) and derived variables based on time series transformed using Symbolic Aggregate Approximation (SAX) with associated time-series pattern cluster membership (Dataset RF_ G) perform the least well with MR of 10.25 and 10.36 respectively. For coronary vascular disease prediction, nearest neighbour (NNge) and the support vector machine based method, SMO, have the highest MR of 10.1 and 10.28 while logistic regression (LR) and the decision tree (DT) method, J48, have MR of 8.85 and 9.0 respectively. DT rules are most comprehensible and clinically relevant. The predictive accuracy increase achieved by addition of risk factor variables to time-series variable based models is significant. The addition of time-series derived variables to models based on risk factor variables alone is associated with a trend to improved performance. Data mining of feature reduced, anaesthesia time-series variables together with risk factor variables can produce compact and moderately accurate models able to predict coronary vascular disease. Decision tree analysis of time-series data combined with risk factor variables yields rules which are more accurate than models based on time-series data alone. The limited additional value provided by electrocardiographic variables when compared to use of risk factors alone is similar to recent suggestions that exercise electrocardiography (exECG) under standardised conditions has limited additional diagnostic value over risk factor analysis and symptom pattern. The effect of the pre-processing used in this study had limited effect when time-series variables and risk factor variables are used as model input. In the absence of risk factor input, the use of time-series variables after outlier removal and time series variables based on physiological variable values’ being outside the accepted normal range is associated with some improvement in model performance.

The Biology, Epidemiology, and Management of Rice Tungro Disease in Asia

Many farmers in South and Southeast Asia describe rice tungro disease as a cancer disease because of the severe damage it causes and the difficulty of controlling it (121). As the most important of the 14 rice viral diseases, tungro was first recognized as a leafhopper-transmitted virus disease in 1963 (88). However, tungro, which means “degenerated growth” in a Filipino dialect, has a much longer history. It is almost certain that tungro was responsible for a disease outbreak that occurred in 1859 in Indonesia, which was referred to at the time as mentek (83). In the past, a variety of names has been given to tungro, including accep na pula in the Philippines, penyakit merah in Malaysia, and yelloworange leaf in Thailand (83).

Socioeconomic position, gender, health behaviours and biomarkers of cardiovascular disease and diabetes

Socio-economic gradients in cardiovascular disease (CVD) and diabetes have been found throughout the developed world and there is some evidence to suggest that these gradients may be steeper for women. Research on social gradients in biological risk factors for CVD and diabetes has received less attention and we do not know the extent to which gradients in biomarkers vary for men and women. We examined the associations between two indicators of socio-economic position (education and household income) and biomarkers of diabetes and cardiovascular disease (CVD) for men and women in a national, population-based study of 11,247 Australian adults. Multi-level linear regression was used to assess associations between education and income and glucose tolerance, dyslipidaemia, blood pressure (BP) and waist circumference before and after adjustment for behaviours (diet, smoking, physical activity, TV viewing time, and alcohol use). Measures of glucose tolerance included fasting plasma glucose and insulin and the results of a glucose tolerance test (2 h glucose) with higher levels of each indicating poorer glucose tolerance. Triglycerides and High Density Lipoprotein (HDL) Cholesterol were used as measures of dyslipidaemia with higher levels of the former and lower levels of the later being associated with CVD risk. Lower education and low income were associated with higher levels of fasting insulin, triglycerides and waist circumference in women. Women with low education had higher systolic and diastolic BP and low income women had higher 2 h glucose and lower HDL cholesterol. With only one exception (low income and systolic BP), all of these estimates were reduced by more than 20% when behavioural risk factors were included. Men with lower education had higher fasting plasma glucose, 2 h glucose, waist circumference and systolic BP and, with the exception of waist circumference, all of these estimates were reduced when health behaviours were included in the models. While low income was associated with higher levels of 2-h glucose and triglycerides it was also associated with better biomarker profiles including lower insulin, waist circumference and diastolic BP. We conclude that low socio-economic position is more consistently associated with a worse profile of biomarkers for CVD and diabetes for women.

Adherence to fish oil intervention in patients with chronic kidney disease

Objective: With growing recognition of the role of inflammation in the development of chronic and acute disease, fish oil is increasingly used as a therapeutic agent, but the nature of the intervention may pose barriers to adherence in clinical populations. Our objective was to investigate the feasibility of using a fish oil supplement in hemodialysis patients. ---------- Design: This was a nonrandomized intervention study.---------- Setting: Eligible patients were recruited at the Hemodialysis Unit of Wesley Hospital, Brisbane, Queensland, Australia. Patients The sample included 28 maintenance hemodialysis patients out of 43 eligible patients in the unit. Exclusion criteria included patients regularly taking a fish oil supplement at baseline, receiving hemodialysis for less than 3 months, or being unable to give informed consent.---------- Intervention: Eicosapentaenoic acid (EPA) was administered at 2000 mg/day (4 capsules) for 12 weeks. Adherence was measured at baseline and weekly throughout the study according to changes in plasma EPA, and was further measured subjectively by self-report.---------- Results: Twenty patients (74%) adhered to the prescription based on changes in plasma EPA, whereas an additional two patients self-reported good adherence. There was a positive relationship between fish oil intake and change in plasma EPA. Most patients did not report problems with taking the fish oil. Using the baseline data, it was not possible to characterize adherent patients.---------- Conclusions: Despite potential barriers, including the need to take a large number of prescribed medications already, 74% of hemodialysis patients adhered to the intervention. This study demonstrated the feasibility of using fish oil in a clinical population.

Compliance, peritoneal dialysis and chronic kidney disease : lessons from the literature

Poor patient compliance with peritoneal dialysis (PD) has significant adverse effects on morbidity and mortality rates in individuals with chronic kidney disease (CKD). It also adds to the resource burdens of healthcare services and providers. This paper explores the notion of PD compliance in patients with CKD with reference to the relevant published literature. The analysis of the literature reveals that ‘PD compliance’ is a complex and challenging construct for both patients and health professionals. There is no universal definition of compliance that is widely adopted in practice and research, and therefore a lack of consensus on how to determine ‘compliant’ patient outcomes. There are also multiple and interconnected determinants of PD compliance that are context-bound, which healthcare professionals must be aware of, and which makes producing consensus of measuring PD compliance difficult. The complexity of the interventions required to produce even a modest improvement in PD compliance, which are described in this paper, are significant. Compliance with PD and other treatments for CKD is a multidimensional, context-bound concept, that to date has tended to efface the role and needs of the renal patient. We conclude the paper with the implications for contemporary practice.

Chronic disease, geographic location and socioeconomic disadvantage as obstacles to equitable access to e-health

Despite recent public attention to e-health as a solution to rising healthcare costs and an ageingpopulation, there have been relatively few studies examining the geographical pattern of e-health usage. This paper argues for an equitable approach to e-health and attention to the way in which e-health initiatives can produce locational health inequalities, particularly in socioeconomically disadvantaged areas. In this paper, we use a case study to demonstrate geographical variation in Internet accessibility, Internet status and prevalence of chronic diseases within a small district. There are signifi cant disparities in access to health information within socioeconomically disadvantaged areas. The most vulnerable people in these areas are likely to have limited availability of, or access to Internet healthcare resources. They are also more likely to have complex chronic diseases and, therefore, be in greatest need of these resources. This case study demonstrates the importance of an equitable approach to e-health information technologies and telecommunications infrastructure.