Alice Richardson

Bounded influence estimation in the mixed linear model

Abstract Bounded influence estimation (also known as generalized M or GM estimation) in the regression model is reviewed. The definitions of bounded influence estimation proposed by Mallows and Schweppe are then extended to the mixed linear model. This is achieved by applying appropriate weight functions to maximum likelihood and restricted maximum likelihood estimating equations. The asymptotic properties of the new estimators are obtained, and the estimators are applied to an artificial dataset. The article concludes with an extension of the example into a small simulation study designed to test some properties of the estimators in samples of moderate size.

Mobile-phone-based classroom response systems: Students’ perceptions of engagement and learning in a large undergraduate course

Using a Classroom Response System (CRS) has been associated with positive educational outcomes, by fostering student engagement and by allowing immediate feedback to both students and instructors. This study examined a low-cost CRS (VotApedia) in a large first-year class, where students responded to questions using their mobile phones. This study explored whether the use of VotApedia retained the advantages of other CRS, overcame some of the challenges of other CRS, and whether new challenges were introduced by using VotApedia. These issues were studied within three themes: students’ perceptions of using VotApedia; the impact of VotApedia on their engagement; and the impact of VotApedia on their learning. Data were collected from an online survey, focus groups and student feedback on teaching and course content. The results indicated that using VotApedia retains the pedagogical advantages of other CRS, while overcoming some of the challenges presented by using other CRS, without introducing any new challenges.

Comorbidity of postural orthostatic tachycardia syndrome and chronic fatigue syndrome in an Australian cohort

Patients with chronic fatigue syndrome (CFS) are frequently diagnosed with comorbid postural orthostatic tachycardia syndrome (POTS), suggesting a shared pathogenesis. The aim of this study was to examine the relationship between demographic characteristics, autonomic functioning and fatigue levels amongst CFS patients with and without comorbid POTS.

An 1H-MRS framework predicts the onset of Alzheimer's disease symptoms in PSEN1 mutation carriers

Alzheimers disease (AD) is the most common cause of dementia; the main risk factors are age and several recently identified genes. A major challenge for AD research is the early detection of subjects at risk. The aim of this study is to develop a predictive model using proton magnetic resonance spectroscopy (1H‐MRS), a noninvasive technique that evaluates brain chemistry in vivo, for monitoring the clinical outcome of carriers of a fully penetrant mutation that causes AD.

Sources of variation in linking corporate social responsibility and financial performance

Purpose - This purpose of this paper is to identify the possible sources of variation of results in prior studies linking corporate social responsibility (CSR) with corporate financial performance (CFP). Design/methodology/approach - A meta-analysis was performed on 51 prior studies included in Orlitzky Findings - The major finding of the study is that sample size and methodology are significant sources of variation in measuring the link between CSR and CFP. Research limitations/implications - The findings are likely to help develop a structural framework towards broadening and deepening our understanding of the debate regarding the sources of variation in the measurement of CSR and CFP link. This research is limited to papers published up to 1999 as included in Orlitzky Originality/value - This paper can be considered an advance on the previous research as it contributes to broadening our understanding of the possible source of causes of variation in results of studies linking CSR with CFP.

Instructor Perceptions of Using a Mobile-Phone-Based Free Classroom Response System in First-Year Statistics Undergraduate Courses

Student engagement at first-year level is critical for student achievement, retention and success. One way of increasing student engagement is to use a classroom response system (CRS), the use of which has been associated with positive educational outcomes by fostering student engagement and by allowing immediate feedback to both students and instructors. Traditional CRS rely on special and often costly hardware (clickers), and often special software, requiring IT support. As a result, the costs of implementation and use may be substantial. This study explores the use of a low-cost CRS (VotApedia) from an instructor perspective. The use of VotApedia enabled first-year students to become anonymously engaged in a large-class environment by using their mobile phones to vote on multiple-choice questions posed by instructors during lectures. VotApedia was used at three Australian universities in first-year undergraduate statistics classes. The instructors in the study collected qualitative and quantitative data specifically related to interacting with the VotApedia interface, the in-class delivery, and instructor perceptions of student engagement. This article presents the instructors’ perceptions of the advantages and challenges of using VotApedia, the practicalities for consideration by potential adopters and recommendations for the future.

Infection status outcome, machine learning method and virus type interact to affect the optimised prediction of hepatitis virus immunoassay results from routine pathology laboratory assays in unbalanced data.

BackgroundAdvanced data mining techniques such as decision trees have been successfully used to predict a variety of outcomes in complex medical environments. Furthermore, previous research has shown that combining the results of a set of individually trained trees into an ensemble-based classifier can improve overall classification accuracy. This paper investigates the effect of data pre-processing, the use of ensembles constructed by bagging, and a simple majority vote to combine classification predictions from routine pathology laboratory data, particularly to overcome a large imbalance of negative Hepatitis B virus (HBV) and Hepatitis C virus (HCV) cases versus HBV or HCV immunoassay positive cases. These methods were illustrated using a never before analysed data set from ACT Pathology (Canberra, Australia) relating to HBV and HCV patients.ResultsIt was easier to predict immunoassay positive cases than negative cases of HBV or HCV. While applying an ensemble-based approach rather than a single classifier had a small positive effect on the accuracy rate, this also varied depending on the virus under analysis. Finally, scaling data before prediction also has a small positive effect on the accuracy rate for this dataset. A graphical analysis of the distribution of accuracy rates across ensembles supports these findings.ConclusionsLaboratories looking to include machine learning as part of their decision support processes need to be aware that the infection outcome, the machine learning method used and the virus type interact to affect the enhanced laboratory diagnosis of hepatitis virus infection, as determined by primary immunoassay data in concert with multiple routine pathology laboratory variables. This awareness will lead to the informed use of existing machine learning methods, thus improving the quality of laboratory diagnosis via informatics analyses.

Predicting the presence of hepatitis B virus surface antigen in Chinese patients by pathology data mining.

Hepatitis B virus (HBV) is a pathogen of worldwide health significance, associated with liver disease. A vaccine is available, yet HBV prevalence remains a concern, particularly in developing countries. Pathology laboratories have a primary role in the diagnosis and monitoring of HBV infection, through hepatitis B surface antigen (HBsAg) immunoassay and associated tests. Analysis of HBsAg immunoassay and associated pathology data from 821 Chinese patients applied 10‐fold cross‐validation to establish classification decision trees (CDTs), with CDT results used subsequently to develop a logistic regression model. The robustness of logistic regression model was confirmed by the Hosmer–Lemeshow test, Pseudo‐R2 and an area under receiver operating characteristic curve (AUROC) result that showed the logistic regression model was capable of accurately discriminating the HBsAg positive from HBsAg negative patients at 95% accuracy. Overall CDT sensitivity and specificity was 94.7% (±5.0%) and 89.5% (±5.7%), respectively, close to the sensitivity and specificity of the immunoassay, providing an alternative to predict HBsAg status. Both the CDT and logistic regression modeling demonstrated the importance of the routine pathology variables alanine aminotransferase (ALT), serum albumin (ALB), and alkaline phosphatase (ALP) to accurately predict HBsAg status in a Chinese patient cohort. The study demonstrates that CDTs and a linked logistic regression model applied to routine pathology data were an effective supplement to HBsAg immunoassay, and a possible replacement method where immunoassays are not requested or not easily available for the laboratory diagnosis of HBV infection. J. Med. Virol. 85:1334–1339, 2013.

An Investigation into the Effect of Ensemble Size and Voting Threshold on the Accuracy of Neural Network Ensembles

If voting is used by an ensemble to classify data, some data points may not be classified, but a higher proportion of those which are classified are classified correctly. This trade off is affected by ensemble size and voting threshold. This paper investigates the effect of ensemble size on the proportions of decisions made and correct decisions. It does this for majority voting and consensus voting on ensembles of neural network classifiers constructed using bagging. It also models the relationships in order to estimate the asymptotic values as the ensemble size increases.

Clinical chemistry in higher dimensions: Machine-learning and enhanced prediction from routine clinical chemistry data

Big Data is having an impact on many areas of research, not the least of which is biomedical science. In this review paper, big data and machine learning are defined in terms accessible to the clinical chemistry community. Seven myths associated with machine learning and big data are then presented, with the aim of managing expectation of machine learning amongst clinical chemists. The myths are illustrated with four examples investigating the relationship between biomarkers in liver function tests, enhanced laboratory prediction of hepatitis virus infection, the relationship between bilirubin and white cell count, and the relationship between red cell distribution width and laboratory prediction of anaemia.