Combining expert opinions in prior elicitation

We consider the problem of combining opinions from different experts in an explicitly model-based way to construct a valid subjective prior in a Bayesian statistical approach. We propose a generic approach by considering a hierarchical model accounting for various sources of variation as well as accounting for potential dependence between experts. We apply this approach to two problems. The first problem deals with a food risk assessment problem involving modelling dose-response for Listeria monocytogenes contamination of mice. Two hierarchical levels of variation are considered (between and within experts) with a complex mathematical situation due to the use of an indirect probit regression. The second concerns the time taken by PhD students to submit their thesis in a particular school. It illustrates a complex situation where three hierarchical levels of variation are modelled but with a simpler underlying probability distribution (log-Normal).

Opportunities and challenges : combining distance and on-campus students in concurrent courses

Through its Electronic Delivery of Gator Engineering (EDGE) program, University of Florida (UF) offers online master’s degrees from participating departments within the UF College of Engineering. Each master’s degree requires 10 courses (3 credit hours each). One interesting and unique aspect of the EDGE pro-gram is that distance learners are registered concurrently in the same courses with traditional on-campus students. This paper examines the specific challenges involved in integrating distance students into on-campus courses, including communication, interaction, access to resources, and equal grading practices.

Solid-state assemblies and optical properties of conjugated oligomers combining fluorene and thiophene units

Two conjugated oligomers, representing elementary segments of fluorene-thiophene copolymers, are compared in terms of the microscopic morphology and the optical properties of thin deposits. The atomic force microscopy morphological data and the solid-state absorption and emission spectra are interpreted in terms of the assembly of the conjugated molecules. The compound with a terthiophene central unit and fluorene end-groups shows well-defined monolayer-by-monolayer assembly into micrometer-long stripe-like structures, with a crystalline herringbone-type organization within the monolayers. Polarized confocal microscopy indicates a strong orientation of the crystalline domains within the stripes. In contrast, the compound with a terfluorene central unit and thiophene end groups forms no textured aggregates and the optical spectra in the solid-state are very similar to those recorded in solution, suggesting that the molecules interact only weakly in the solid. The difference in behaviour between the two compounds most probably originates from their different capability to form densely-packed assemblies of interacting π-systems.

Combining passive sampling and toxicity testing for evaluation of mixtures of polar organic chemicals in sewage treatment plant effluent

Effluent from sewage treatment plants has been associated with a range of pollutant effects. Depending on the influent composition and treatment processes the effluent may contain a myriad of different chemicals which makes monitoring very complex. In this study we aimed to monitor relatively polar organic pollutant mixtures using a combination of passive sampling techniques and a set of biochemistry based assays covering acute bacterial toxicity (Microtox™), phytotoxicity (Max-I-PAM assay) and genotoxicity (umuC assay). The study showed that all of the assays were able to detect effects in the samples and allowed a comparison of the two plants as well as a comparison between the two sampling periods. Distinct improvements in water quality were observed in one of the plants as result of an upgrade to a UV disinfection system, which improved from 24× sample enrichment required to induce a 50% response in the Microtox™ assay to 84×, from 30× sample enrichment to induce a 50% reduction in photosynthetic yield to 125×, and the genotoxicity observed in the first sampling period was eliminated. Thus we propose that biochemical assay techniques in combination with time integrated passive sampling can substantially contribute to the monitoring of polar organic toxicants in STP effluents.

Examining the predictive value of combining the theory of planned behaviour and the volunteer functions inventory

The functions of the volunteer functions inventory were combined with the constructs of the theory of planned behaviour (i.e., attitudes, subjective norms, and perceived behavioural control) to establish whether a stronger, single explanatory model prevailed. Undertaken in the context of episodic, skilled volunteering by individuals who were retired or approaching retirement (N = 186), the research advances on prior studies which either examined the predictive capacity of each model independently or compared their explanatory value. Using hierarchical regression analysis, the functions of the volunteer functions inventory (when controlling for demographic variables) explained an additional 7.0% of variability in individuals’ willingness to volunteer over and above that accounted for by the theory of planned behaviour. Significant predictors in the final model included attitudes, subjective norms and perceived behavioural control from the theory of planned behaviour and the understanding function from the volunteer functions inventory. It is proposed that the items comprising the understanding function may represent a deeper psychological construct (e.g., self-actualisation) not accounted for by the theory of planned behaviour. The findings highlight the potential benefit of combining these two prominent models in terms of improving understanding of volunteerism and providing a single parsimonious model for raising rates of this important behaviour.

Combining opinions for use in Bayesian networks: A measurement error approach

Bayesian networks (BNs) are graphical probabilistic models used for reasoning under uncertainty. These models are becoming increasing popular in a range of fields including ecology, computational biology, medical diagnosis, and forensics. In most of these cases, the BNs are quantified using information from experts, or from user opinions. An interest therefore lies in the way in which multiple opinions can be represented and used in a BN. This paper proposes the use of a measurement error model to combine opinions for use in the quantification of a BN. The multiple opinions are treated as a realisation of measurement error and the model uses the posterior probabilities ascribed to each node in the BN which are computed from the prior information given by each expert. The proposed model addresses the issues associated with current methods of combining opinions such as the absence of a coherent probability model, the lack of the conditional independence structure of the BN being maintained, and the provision of only a point estimate for the consensus. The proposed model is applied an existing Bayesian Network and performed well when compared to existing methods of combining opinions.

Combining initial segments of lists

We propose a new way to build a combined list from K base lists, each containing N items. A combined list consists of top segments of various sizes from each base list so that the total size of all top segments equals N. A sequence of item requests is processed and the goal is to minimize the total number of misses. That is, we seek to build a combined list that contains all the frequently requested items. We first consider the special case of disjoint base lists. There, we design an efficient algorithm that computes the best combined list for a given sequence of requests. In addition, we develop a randomized online algorithm whose expected number of misses is close to that of the best combined list chosen in hindsight. We prove lower bounds that show that the expected number of misses of our randomized algorithm is close to the optimum. In the presence of duplicate items, we show that computing the best combined list is NP-hard. We show that our algorithms still apply to a linearized notion of loss in this case. We expect that this new way of aggregating lists will find many ranking applications.

Prior and posterior linear pooling for combining expert opinions: Uses and impact on Bayesian Networks

The use of expert knowledge to quantify a Bayesian Network (BN) is necessary when data is not available. This however raises questions regarding how opinions from multiple experts can be used in a BN. Linear pooling is a popular method for combining probability assessments from multiple experts. In particular, Prior Linear Pooling (PrLP), which pools opinions then places them into the BN is a common method. This paper firstly proposes an alternative pooling method, Posterior Linear Pooling (PoLP). This method constructs a BN for each expert, then pools the resulting probabilities at the nodes of interest. Secondly, it investigates the advantages and disadvantages of using these pooling methods to combine the opinions of multiple experts. Finally, the methods are applied to an existing BN, the Wayfinding Bayesian Network Model, to investigate the behaviour of different groups of people and how these different methods may be able to capture such differences. The paper focusses on 6 nodes Human Factors, Environmental Factors, Wayfinding, Communication, Visual Elements of Communication and Navigation Pathway, and three subgroups Gender (female, male),Travel Experience (experienced, inexperienced), and Travel Purpose (business, personal) and finds that different behaviors can indeed be captured by the different methods.

Additive tissue manufacturing for breast reconstruction: Combining CAD/CAM with adipose tissue engineering

The primary aim of this multidisciplinary project was to develop a new generation of breast implants. Disrupting the currently prevailing paradigm of silicone implants which permanently introduce a foreign body into mastectomy patients, highly porous implants developed as part of this PhD project are biodegradable by the body and augment the growth of natural tissue. Our technology platform leverages computer-assisted-design which allows us to manufacture fully patient-specific implants based on a personalised medicine approach. Multiple animal studies conducted in this project have shown that the polymeric implant slowly degrades within the body harmlessly while the body's own tissue forms concurrently.

Combining meta- and mega-analytic approaches for multi-site diffusion imaging based genetic studies: From the enigma-DTI working group

Meta-analyses estimate a statistical effect size for a test or an analysis by combining results from multiple studies without necessarily having access to each individual study's raw data. Multi-site meta-analysis is crucial for imaging genetics, as single sites rarely have a sample size large enough to pick up effects of single genetic variants associated with brain measures. However, if raw data can be shared, combining data in a "mega-analysis" is thought to improve power and precision in estimating global effects. As part of an ENIGMA-DTI investigation, we use fractional anisotropy (FA) maps from 5 studies (total N=2, 203 subjects, aged 9-85) to estimate heritability. We combine the studies through meta-and mega-analyses as well as a mixture of the two - combining some cohorts with mega-analysis and meta-analyzing the results with those of the remaining sites. A combination of mega-and meta-approaches may boost power compared to meta-analysis alone.

Combining transcriptome assemblies from multiple de novo assemblers in the allo-tetraploid plant Nicotiana benthamiana

Background Nicotiana benthamiana is an allo-tetraploid plant, which can be challenging for de novo transcriptome assemblies due to homeologous and duplicated gene copies. Transcripts generated from such genes can be distinct yet highly similar in sequence, with markedly differing expression levels. This can lead to unassembled, partially assembled or mis-assembled contigs. Due to the different properties of de novo assemblers, no one assembler with any one given parameter space can re-assemble all possible transcripts from a transcriptome. Results In an effort to maximise the diversity and completeness of de novo assembled transcripts, we utilised four de novo transcriptome assemblers, TransAbyss, Trinity, SOAPdenovo-Trans, and Oases, using a range of k-mer sizes and different input RNA-seq read counts. We complemented the parameter space biologically by using RNA from 10 plant tissues. We then combined the output of all assemblies into a large super-set of sequences. Using a method from the EvidentialGene pipeline, the combined assembly was reduced from 9.9 million de novo assembled transcripts to about 235,000 of which about 50,000 were classified as primary. Metrics such as average bit-scores, feature response curves and the ability to distinguish paralogous or homeologous transcripts, indicated that the EvidentialGene processed assembly was of high quality. Of 35 RNA silencing gene transcripts, 34 were identified as assembled to full length, whereas in a previous assembly using only one assembler, 9 of these were partially assembled. Conclusions To achieve a high quality transcriptome, it is advantageous to implement and combine the output from as many different de novo assemblers as possible. We have in essence taking the ‘best’ output from each assembler while minimising sequence redundancy. We have also shown that simultaneous assessment of a variety of metrics, not just focused on contig length, is necessary to gauge the quality of assemblies.

Tracking the patient journey by combining multiple hospital database systems

With new national targets for patient flow in public hospitals designed to increase efficiencies in patient care and resource use, better knowledge of events affecting length of stay will support improved bed management and scheduling of procedures. This paper presents a case study involving the integration of material from each of three databases in operation at one tertiary hospital and demonstrates it is possible to follow patient journeys from admission to discharge. What is known about this topic? At present, patient data at one Queensland tertiary hospital are assembled in three information systems: (1) the Hospital Based Corporate Information System (HBCIS), which tracks patients from in-patient admission to discharge; (2) the Emergency Department Information System (EDIS) containing patient data from presentation to departure from the emergency department; and (3) Operation Room Management Information System (ORMIS), which records surgical operations. What does this paper add? This paper describes how a new enquiry tool may be used to link the three hospital information systems for studying the hospital journey through different wards and/or operating theatres for both individual and groups of patients. What are the implications for practitioners? An understanding of the patients’ journeys provides better insight into patient flow and provides the tool for research relating to access block, as well as optimising the use of physical and human resources.

Combining HPLC-DAD and ICP-MS data for improved analysis of complex samples: Classification of the root samples from Cortex moutan

A combined data matrix consisting of high performance liquid chromatography–diode array detector (HPLC–DAD) and inductively coupled plasma-mass spectrometry (ICP-MS) measurements of samples from the plant roots of the Cortex moutan (CM), produced much better classification and prediction results in comparison with those obtained from either of the individual data sets. The HPLC peaks (organic components) of the CM samples, and the ICP-MS measurements (trace metal elements) were investigated with the use of principal component analysis (PCA) and the linear discriminant analysis (LDA) methods of data analysis; essentially, qualitative results suggested that discrimination of the CM samples from three different provinces was possible with the combined matrix producing best results. Another three methods, K-nearest neighbor (KNN), back-propagation artificial neural network (BP-ANN) and least squares support vector machines (LS-SVM) were applied for the classification and prediction of the samples. Again, the combined data matrix analyzed by the KNN method produced best results (100% correct; prediction set data). Additionally, multiple linear regression (MLR) was utilized to explore any relationship between the organic constituents and the metal elements of the CM samples; the extracted linear regression equations showed that the essential metals as well as some metallic pollutants were related to the organic compounds on the basis of their concentrations