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.

Examination of the latent structure of the psychological sense of school membership scale

Despite its widespread use, there has been limited examination of the underlying factor structure of the Psychological Sense of School Membership (PSSM) scale. The current study examined the psychometric properties of the PSSM to refine its utility for researchers and practitioners using a sample of 504 Australian high school students. Results from exploratory and confirmatory factor analyses indicated that the PSSM is a multidimensional instrument. Factor analysis procedures identified three factors representing related aspects of students’ perceptions of their school membership: caring relationships, acceptance, and rejection

Validation of the Cannabis Expectancy Questionnaire (CEQ) in adult cannabis users in treatment

Background Outcome expectancies are a key cognitive construct in the etiology, assessment and treatment of Substance Use Disorders. There is a research and clinical need for a cannabis expectancy measure validated in a clinical sample of cannabis users. Method The Cannabis Expectancy Questionnaire (CEQ) was subjected to exploratory (n = 501, mean age 27.45, 78% male) and confirmatory (n = 505, mean age 27.69, 78% male) factor analysis in two separate samples of cannabis users attending an outpatient cannabis treatment program. Weekly cannabis consumption was clinically assessed and patients completed the Severity of Dependence Scale-Cannabis (SDS-C) and the General Health Questionnaire (GHQ-28). Results Two factors representing Negative Cannabis Expectancies and Positive Cannabis Expectancies were identified. These provided a robust statistical and conceptual fit for the data. Internal reliabilities were high. Negative expectancies were associated with greater dependence severity (as measured by the SDS) and positive expectancies with higher consumption. The interaction of positive and negative expectancies was consistently significantly associated with self-reported functioning across all four GHQ-28 scales (Somatic Concerns, Anxiety, Social Dysfunction and Depression). Specifically, within the context of high positive cannabis expectancy, higher negative expectancy was predictive of more impaired functioning. By contrast, within the context of low positive cannabis expectancy, higher negative expectancy was predictive of better functioning. Conclusions The CEQ is the first cannabis expectancy measure to be validated in a sample of cannabis users in treatment. Negative and positive cannabis expectancy domains were uniquely associated with consumption, dependence severity and self-reported mental health functioning.

An examination of the factorial invariance and refinement of the multidimensional school anger inventory for five Pacific Rim countries

The validity of the Multidimensional School Anger Inventory (MSAI) was examined with adolescents from 5 Pacific Rim countries (N ¼ 3,181 adolescents; age, M ¼ 14.8 years; 52% females). Confirmatory factor analyses examined configural invariance for the MSAI’s anger experience, hostility, destructive expression, and anger coping subscales. The model did not converge for Peruvian students. Using the top 4 loaded items for anger experience, hostility, and destructive expression configural invariance and partial metric and scalar invariances were found. Latent means analysis compared mean responses on each subscale to the U.S. sample. Students from other countries showed higher mean responses on the anger experience subscale (ds ¼ .37–.73). Australian (d ¼ .40) and Japanese students (d ¼ .21) had significantly higher mean hostility subscale scores. Australian students had higher mean scores on the destructive expression subscale (d ¼ .30), whereas Japanese students had lower mean scores (d ¼ 2.17). The largest latent mean gender differences (females lower than males) were for destructive expression among Australian (d ¼ 2.67), Guatemalan (d ¼ 2.42), and U.S. (d ¼ 2.66) students. This study supported an abbreviated, 12-item MSAI with partial invariance. Implications for the use of the MSAI in comparative research are discussed.

Designing the information architecture of governmental one-stop portals : on the application and analysis of card sorting

The latest paradigm shift in government, termed Transformational Government, puts the citizen in the centre of attention. Including citizens in the design of online one-stop portals can help governmental organisations to become more customer focussed. This study describes the initial efforts of an Australian state government to develop an information architecture to structure the content of their future one-stop portal. Hereby, card sorting exercises have been conducted and analysed, utilising contemporary approaches found in academic and non-scientific literature. This paper describes the findings of the card sorting exercises in this particular case and discusses the suitability of the applied approaches in general. These are distinguished into non-statistical, statistical, and hybrid approaches. Thus, on the one hand, this paper contributes to academia by describing the application of different card sorting approaches and discussing their strengths and weaknesses. On the other hand, this paper contributes to practice by explaining the approach that has been taken by the authors’ research partner in order to develop a customer-focussed governmental one-stop portal. Thus, they provide decision support for practitioners with regard to different analysis methods that can be used to complement recent approaches in Transformational Government.

Brief screening for psychosis-like experiences

Objective Psychotic-like experiences (PLEs) are common, and are markers of poor mental health. This study examined the internal structure of a screening test, the Community Assessment of Psychic Experiences-Positive scale (CAPE-P) in a young Australian sample. Method A cross-sectional online survey, which included the CAPE-P, was completed by 1610 university students aged between 18 and 25 years. Confirmatory factor analyses compared 1-, 4-, and 5-factor models, and examined effects of omitting selected items. Results A 3-factor model, omitting items on magical thinking, grandiosity, paranormal beliefs and a cross-loading item produced the best fit. The resultant 15-item CAPE (CAPE-P15) had three subscales - Persecutory Ideation, Perceptual Abnormalities and Bizarre Experiences, all with high levels of internal consistency. Conclusion The CAPE-P15 shows promise as a measure of positive, psychosis-like experiences, but further validation of this measure is required in community samples.

Psychometric validation of the Dysexecutive Questionnaire (DEX)

This study reported on the validation of the psychometric properties, the factorability, validity, and sensitivity of the Dysexecutive Questionnaire (DEX) in 3 clinical and nonclinical samples. A mixed sample of 997 participants—community (n = 663), psychiatric (depressed [n = 92] and anxious [n = 122]), and neurologically impaired (n = 120)—completed self-report questionnaires assessing executive dysfunction, depression, anxiety, stress, general self-efficacy, and satisfaction with life. Before analyses the data were randomly split into 2 subsets (A and B). Exploratory factor analysis performed on Subset A produced a 3-factor model (Factor 1: Inhibition, Factor 2: Volition, and Factor 3: Social Regulation) in which 15 of the original 20 items provided a revised factor structure that was superior to all other structures. A series of confirmatory factor analyses performed on Subset B confirmed that this revised factor structure was valid and reliable. The revised structure, labeled the DEX-R, was found to be a reliable and valid tool for assessing behavioral symptoms of dysexecutive functioning in mixed community, psychiatric, and neurological samples.

The Governance of human resources in the Vietnamese healthcare system: A critical analysis of maternity services

This thesis utilised mixed-methods study design to understand the factors that influence the translation and implementation of central human resources in health policy at the district and commune health levels. It provided recommendations for changes to enhance governance approaches to human resources for health policy implementation at local and national levels. This thesis has also contributed to the evolution of the theory on health staff motivation and performance through the description and testing of a new model, using data from a survey on 262 health staff and 43 in-depth interviews conducted in two northern mountainous provinces of Vietnam.

Analysis of complex molecular systems: The impact of multivariate analysis for resolving the interactions of small molecules with biopolymers - a review

This review is focused on the impact of chemometrics for resolving data sets collected from investigations of the interactions of small molecules with biopolymers. These samples have been analyzed with various instrumental techniques, such as fluorescence, ultraviolet–visible spectroscopy, and voltammetry. The impact of two powerful and demonstrably useful multivariate methods for resolution of complex data—multivariate curve resolution–alternating least squares (MCR–ALS) and parallel factor analysis (PARAFAC)—is highlighted through analysis of applications involving the interactions of small molecules with the biopolymers, serum albumin, and deoxyribonucleic acid. The outcomes illustrated that significant information extracted by the chemometric methods was unattainable by simple, univariate data analysis. In addition, although the techniques used to collect data were confined to ultraviolet–visible spectroscopy, fluorescence spectroscopy, circular dichroism, and voltammetry, data profiles produced by other techniques may also be processed. Topics considered including binding sites and modes, cooperative and competitive small molecule binding, kinetics, and thermodynamics of ligand binding, and the folding and unfolding of biopolymers. Applications of the MCR–ALS and PARAFAC methods reviewed were primarily published between 2008 and 2013.

Classifying patients by their characteristics and clinical presentations; the use of latent class analysis

In this article, we introduce the general statistical analysis approach known as latent class analysis and discuss some of the issues associated with this type of analysis in practice. Two recent examples from the respiratory health literature are used to highlight the types of research questions that have been addressed using this approach.

The role of moral disengagement in underage drinking and heavy episodic drinking

Objectives The current study had two aims. First, to develop a moral disengagement scale contextualized to underage drinking. Second, to investigate Bandura’s (1986) self-regulatory model within the context of underage drinking. Method Two different samples of students participated in the study. The first sample included 619 (362 females) adolescents (Mage = 15.3 years, SD = 1.09 years) and the second sample 636 (386 females) adolescents (Mage = 15.3 years, SD = 1.03 years). Students in the first sample completed the Underage Drinking Disengagement Scale (UDDS), and measures of engagement in underage drinking and heavy episodic drinking. Students in the second sample completed these measures as well as scales of general moral disengagement, personal standards and anticipatory guilt associated with underage drinking. Results For the UDDS, exploratory and confirmatory factor analyses verified a single factor structure. The UDDS was more strongly associated with engagement in underage drinking and heavy episodic drinking than a general measure of moral disengagement. A moderated mediation analysis revealed that adolescents who negatively evaluated underage drinking reported more anticipatory guilt, and more anticipatory guilt was associated with less engagement in underage drinking and less heavy episodic drinking. This relationship was weaker at high compared to low levels of underage drinking disengagement. Conclusions/Importance Understanding how adolescents self-regulate their drinking, and ways that such self-regulation may be deactivated or disengaged, may help identify those adolescents at increased risk of drinking underage and of engaging in heavy episodic drinking.

Knowledge management for ERP success

The business value of Enterprise Resource Planning (ERP) systems and in general large software implementations has been extensively debated in both popular press and academic literature for over three decades. Despite the positive motives for adoption, various organizations have reported negative impacts from these large investments. This ‘disconnect’ between large IS investments and firms’ organizational performance may be attributable to the economic transition from an era of competitive advantage based on information to one that is based on Knowledge. This paper discusses the initial findings of a two-phased study that focuses on empirically assessing the impact of knowledge management on the success of Enterprise Resource Planning systems. The research study uses information gathered from twenty-seven public sector organizations in Queensland, Australia. Validation of the a priori model constructs through factor analysis identified two dimensions of knowledge management. Further analysis assessed the comparative differences in perceptions of knowledge management in ERP, across four employment cohorts.

Stability of the learning and development survey : findings for mature aged local government and private healthcare organisations.

This article investigates work related learning and development amongst mature aged workers from a lifespan developmental psychology perspective. The current study follows on from research regarding the construction and revision of the Learning and Development Survey (LDS; Tones & Pillay, 2008). Designed to measure adaptive development for work related learning, the revised LDS (R-LDS) encompasses goal selection, engagement and disengagement from individual and organisational perspectives. Previous survey findings from a mixed age sample of local government workers suggest that mature aged workers aged over 45 years are less likely to report engagement in learning and development goals than younger workers, which is partly due to insufficient opportunities at work. In the current paper, exploratory factor analysis was used to investigate responses to the R-LDS amongst two groups of mature aged workers from a local government (LG) and private healthcare (PH) organisation to determine the stability of the R-LDS. Organisational constraints to development accounted for almost a quarter of the variance in R-LDS scores for both samples, while remaining factors emerged in different orders for each data set. Organisational opportunities for development explained about 17% of the variance in R-LDS scores in the LG sample, while the individual goal disengagement factor contributed a comparable proportion of variance to R-LDS scores for the PH sample. Findings from the current study indicate that opportunities for learning and development at work may be age structured and biased towards younger workers. Implications for professional practice are discussed and focus on improving the engagement of mature aged workers.

Developing an integrative model for understanding innovation adoption

We develop and test a theoretically-based integrative model of organizational innovation adoption. Confirmatory factor analyses using responses from 134 organizations showed that the hypothesized second-order model was a better fit to the data than the traditional model of independent factors. Furthermore, although not all elements were significant, the hypothesized model fit adoption better than the traditional model.