Eliciting and encoding expert knowledge on variable selection into classical or Bayesian species distribution models

The quality of species distribution models (SDMs) relies to a large degree on the quality of the input data, from bioclimatic indices to environmental and habitat descriptors (Austin, 2002). Recent reviews of SDM techniques, have sought to optimize predictive performance e.g. Elith et al., 2006. In general SDMs employ one of three approaches to variable selection. The simplest approach relies on the expert to select the variables, as in environmental niche models Nix, 1986 or a generalized linear model without variable selection (Miller and Franklin, 2002). A second approach explicitly incorporates variable selection into model fitting, which allows examination of particular combinations of variables. Examples include generalized linear or additive models with variable selection (Hastie et al. 2002); or classification trees with complexity or model based pruning (Breiman et al., 1984, Zeileis, 2008). A third approach uses model averaging, to summarize the overall contribution of a variable, without considering particular combinations. Examples include neural networks, boosted or bagged regression trees and Maximum Entropy as compared in Elith et al. 2006. Typically, users of SDMs will either consider a small number of variable sets, via the first approach, or else supply all of the candidate variables (often numbering more than a hundred) to the second or third approaches. Bayesian SDMs exist, with several methods for eliciting and encoding priors on model parameters (see review in Low Choy et al. 2010). However few methods have been published for informative variable selection; one example is Bayesian trees (O’Leary 2008). Here we report an elicitation protocol that helps makes explicit a priori expert judgements on the quality of candidate variables. This protocol can be flexibly applied to any of the three approaches to variable selection, described above, Bayesian or otherwise. We demonstrate how this information can be obtained then used to guide variable selection in classical or machine learning SDMs, or to define priors within Bayesian SDMs.

Mathematical models of calcium and tight junctions in normal and reconstructed epidermis

This project investigated the calcium distributions of the skin, and the growth patterns of skin substitutes grown in the laboratory, using mathematical models. The research found that the calcium distribution in the upper layer of the skin is controlled by three different mechanisms, not one as previously thought. The research also suggests that tight junctions, which are adhesions between neighbouring skin cells, cannot be solely responsible for the differences in the growth patterns of skin substitutes and normal skin.

Product-library-derived specification parameters in building information models augmentation

This paper conceptualizes a framework for bridging the BIM (building information modelling)-specifications divide through augmenting objects within BIM with specification parameters derived from a product library. We demonstrate how model information, enriched with data at various LODs (levels of development), can evolve simultaneously with design and construction using different representation of a window object embedded in a wall as lifecycle phase exemplars at different levels of granularity. The conceptual standpoint is informed by the need for exploring a methodological approach which extends beyond current limitations of current modelling platforms in enhancing the information content of BIM models. Therefore, this work demonstrates that BIM objects can be augmented with construction specification parameters leveraging product libraries.

Paraxial schematic eye models for 7- and 14-year-old Chinese children

Purpose.: To develop three-surface paraxial schematic eyes with different ages and sexes based on data for 7- and 14-year-old Chinese children from the Anyang Childhood Eye Study. Methods.: Six sets of paraxial schematic eyes, including 7-year-old eyes, 7-year-old male eyes, 7-year-old female eyes, 14-year-old eyes, 14-year-old male eyes, and 14-year-old female eyes, were developed. Both refraction-dependent and emmetropic eye models were developed, with the former using linear dependence of ocular parameters on refraction. Results.: A total of 2059 grade 1 children (boys 58%) and 1536 grade 8 children (boys 49%) were included, with mean age of 7.1 ± 0.4 and 13.7 ± 0.5 years, respectively. Changes in these schematic eyes with aging are increased anterior chamber depth, decreased lens thickness, increased vitreous chamber depth, increased axial length, and decreased lens equivalent power. Male schematic eyes have deeper anterior chamber depth, longer vitreous chamber depth, longer axial length, and lower lens equivalent power than female schematic eyes. Changes in the schematic eyes with positive increase in refraction are decreased anterior chamber depth, increased lens thickness, decreased vitreous chamber depth, decreased axial length, increased corneal radius of curvature, and increased lens power. In general, the emmetropic schematic eyes have biometric parameters similar to those arising from regression fits for the refraction-dependent schematic eyes. Conclusions.: The paraxial schematic eyes of Chinese children may be useful for myopia research and for facilitating comparison with other children with the same or different racial backgrounds and living in different places.

Monitoring healing progression and characterizing the mechanical environment in preclinical models for bone tissue engineering

The treatment of large segmental bone defects remains a significant clinical challenge. Due to limitations surrounding the use of bone grafts, tissue-engineered constructs for the repair of large bone defects could offer an alternative. Before translation of any newly developed tissue engineering (TE) approach to the clinic, efficacy of the treatment must be shown in a validated preclinical large animal model. Currently, biomechanical testing, histology, and microcomputed tomography are performed to assess the quality and quantity of the regenerated bone. However, in vivo monitoring of the progression of healing is seldom performed, which could reveal important information regarding time to restoration of mechanical function and acceleration of regeneration. Furthermore, since the mechanical environment is known to influence bone regeneration, and limb loading of the animals can poorly be controlled, characterizing activity and load history could provide the ability to explain variability in the acquired data sets and potentially outliers based on abnormal loading. Many approaches have been devised to monitor the progression of healing and characterize the mechanical environment in fracture healing studies. In this article, we review previous methods and share results of recent work of our group toward developing and implementing a comprehensive biomechanical monitoring system to study bone regeneration in preclinical TE studies.

Evaluation of k-epsilon and RNG turbulence models for confined swirling and non-swirling flow

In a very recent study [1] the Renormalisation Group (RNG) turbulence model was used to obtain flow predictions in a strongly swirling quarl burner, and was found to perform well in predicting certain features that are not well captured using less sophisticated models of turbulence. The implication is that the RNG approach should provide an economical and reliable tool for the prediction of swirling flows in combustor and furnace geometries commonly encountered in technological applications. To test this hypothesis the present work considers flow in a model furnace for which experimental data is available [2]. The essential features of the flow which differentiate it from the previous study [1] are that the annular air jet entry is relatively narrow and the base wall of the cylindrical furnace is at 90 degrees to the inlet pipe. For swirl numbers of order 1 the resulting flow is highly complex with significant inner and outer recirculation regions. The RNG and standard k-epsilon models are used to model the flow for both swirling and non-swirling entry jets and the results compared with experimental data [2]. Near wall viscous effects are accounted for in both models via the standard wall function formulation [3]. For the RNG model, additional computations with grid placement extending well inside the near wall viscous-affected sublayer are performed in order to assess the low Reynolds number capabilities of the model.

Computation of confined swirling flow using RNG and k-epsilon turbulence models

In this work we numerically model isothermal turbulent swirling flow in a cylindrical burner. Three versions of the RNG k-epsilon model are assessed against performance of the standard k-epsilon model. Sensitivity of numerical predictions to grid refinement, differing convective differencing schemes and choice of (unknown) inlet dissipation rate, were closely scrutinised to ensure accuracy. Particular attention is paid to modelling the inlet conditions to within the range of uncertainty of the experimental data, as model predictions proved to be significantly sensitive to relatively small changes in upstream flow conditions. We also examine the characteristics of the swirl--induced recirculation zone predicted by the models over an extended range of inlet conditions. Our main findings are: - (i) the standard k-epsilon model performed best compared with experiment; - (ii) no one inlet specification can simultaneously optimize the performance of the models considered; - (iii) the RNG models predict both single-cell and double-cell IRZ characteristics, the latter both with and without additional internal stagnation points. The first finding indicates that the examined RNG modifications to the standard k-e model do not result in an improved eddy viscosity based model for the prediction of swirl flows. The second finding suggests that tuning established models for optimal performance in swirl flows a priori is not straightforward. The third finding indicates that the RNG based models exhibit a greater variety of structural behaviour, despite being of the same level of complexity as the standard k-e model. The plausibility of the predicted IRZ features are discussed in terms of known vortex breakdown phenomena.

Concise review: Humanized models of tumor immunology in the 21st century: Convergence of cancer research and tissue engineering

Despite positive testing in animal studies, more than 80% of novel drug candidates fail to proof their efficacy when tested in humans. This is primarily due to the use of preclinical models that are not able to recapitulate the physiological or pathological processes in humans. Hence, one of the key challenges in the field of translational medicine is to “make the model organism mouse more human.” To get answers to questions that would be prognostic of outcomes in human medicine, the mouse's genome can be altered in order to create a more permissive host that allows the engraftment of human cell systems. It has been shown in the past that these strategies can improve our understanding of tumor immunology. However, the translational benefits of these platforms have still to be proven. In the 21st century, several research groups and consortia around the world take up the challenge to improve our understanding of how to humanize the animal's genetic code, its cells and, based on tissue engineering principles, its extracellular microenvironment, its tissues, or entire organs with the ultimate goal to foster the translation of new therapeutic strategies from bench to bedside. This article provides an overview of the state of the art of humanized models of tumor immunology and highlights future developments in the field such as the application of tissue engineering and regenerative medicine strategies to further enhance humanized murine model systems.

Multi-GNSS precise point positioning with raw single-frequency and dual-frequency measurement models

The emergence of multiple satellite navigation systems, including BDS, Galileo, modernized GPS, and GLONASS, brings great opportunities and challenges for precise point positioning (PPP). We study the contributions of various GNSS combinations to PPP performance based on undifferenced or raw observations, in which the signal delays and ionospheric delays must be considered. A priori ionospheric knowledge, such as regional or global corrections, strengthens the estimation of ionospheric delay parameters. The undifferenced models are generally more suitable for single-, dual-, or multi-frequency data processing for single or combined GNSS constellations. Another advantage over ionospheric-free PPP models is that undifferenced models avoid noise amplification by linear combinations. Extensive performance evaluations are conducted with multi-GNSS data sets collected from 105 MGEX stations in July 2014. Dual-frequency PPP results from each single constellation show that the convergence time of undifferenced PPP solution is usually shorter than that of ionospheric-free PPP solutions, while the positioning accuracy of undifferenced PPP shows more improvement for the GLONASS system. In addition, the GLONASS undifferenced PPP results demonstrate performance advantages in high latitude areas, while this impact is less obvious in the GPS/GLONASS combined configuration. The results have also indicated that the BDS GEO satellites have negative impacts on the undifferenced PPP performance given the current “poor” orbit and clock knowledge of GEO satellites. More generally, the multi-GNSS undifferenced PPP results have shown improvements in the convergence time by more than 60 % in both the single- and dual-frequency PPP results, while the positioning accuracy after convergence indicates no significant improvements for the dual-frequency PPP solutions, but an improvement of about 25 % on average for the single-frequency PPP solutions.

Genetic analysis of structural brain connectivity using DICCCOL models of diffusion MRI in 522 twins

Genetic and environmental factors affect white matter connectivity in the normal brain, and they also influence diseases in which brain connectivity is altered. Little is known about genetic influences on brain connectivity, despite wide variations in the brain's neural pathways. Here we applied the 'DICCCOL' framework to analyze structural connectivity, in 261 twin pairs (522 participants, mean age: 21.8 y ± 2.7SD). We encoded connectivity patterns by projecting the white matter (WM) bundles of all 'DICCCOLs' as a tracemap (TM). Next we fitted an A/C/E structural equation model to estimate additive genetic (A), common environmental (C), and unique environmental/error (E) components of the observed variations in brain connectivity. We found 44 'heritable DICCCOLs' whose connectivity was genetically influenced (α2>1%); half of them showed significant heritability (α2>20%). Our analysis of genetic influences on WM structural connectivity suggests high heritability for some WM projection patterns, yielding new targets for genome-wide association studies.

How can evaluative thinking and practice inform performance models for organizations and personnel

In the world today there are many ways in which we measure, count and determine whether something is worth the effort or not. In Australia and many other countries, new government legislation is requiring government-funded entities to become more transparent in their practice and to develop a more cohesive narrative about the worth, or impact, for the betterment of society. This places the executives of such entities in a position of needing evaluative thinking and practice to guide how they may build the narrative that documents and demonstrates this type of impact. In thinking about where to start, executives, project and program managers may consider this workshop as a professional development opportunity to explore both the intended and unintended consequences of performance models as tools of evaluation. This workshop will offer participants an opportunity to unpack the place of performance models as an evaluative tool through the following: · What shape does an ethical, sound and valid performance measure for an organization or personnel take? · What role does cultural specificity play in the design and development of a performance model for an organization or for personnel? · How are stakeholders able to identify risk during the design and development of such models? · When and where will dissemination strategies be required? · And so what? How can you determine that your performance model implementation has made a difference now or in the future?

Automated Discovery of Structured Process Models: Discover Structured vs. Discover and Structure

This paper addresses the problem of discovering business process models from event logs. Existing approaches to this problem strike various tradeoffs between accuracy and understandability of the discovered models. With respect to the second criterion, empirical studies have shown that block-structured process models are generally more understandable and less error-prone than unstructured ones. Accordingly, several automated process discovery methods generate block-structured models by construction. These approaches however intertwine the concern of producing accurate models with that of ensuring their structuredness, sometimes sacrificing the former to ensure the latter. In this paper we propose an alternative approach that separates these two concerns. Instead of directly discovering a structured process model, we first apply a well-known heuristic technique that discovers more accurate but sometimes unstructured (and even unsound) process models, and then transform the resulting model into a structured one. An experimental evaluation shows that our “discover and structure” approach outperforms traditional “discover structured” approaches with respect to a range of accuracy and complexity measures.

(Failed) voluntary euthanasia law reform in Australia: Two decades of trends, models and politics

Within Australia, there have been many attempts to pass voluntary euthanasia (VE) or physician-assisted suicide (PAS) legislation. From 16 June 1993 until the date of writing, 51 Bills have been introduced into Australian parliaments dealing with legalising VE or PAS. Despite these numerous attempts, the only successful Bill was the Rights of the Terminally Ill Act 1995 (NT), which was enacted in the Northern Territory, but a short time later overturned by the controversial Euthanasia Laws Act 1997 (Cth). Yet, in stark contrast to the significant political opposition, for decades Australian public opinion has overwhelmingly supported law reform legalising VE or PAS. While there is ongoing debate in Australia, both through public discourse and scholarly publications, about the merits and dangers of reform in this field, there has been remarkably little analysis of the numerous legislative attempts to reform the law, and the context in which those reform attempts occurred. The aim of this article is to better understand the reform landscape in Australia over the past two decades. The information provided in this article will better equip Australians, both politicians and the general public, to have a more nuanced understanding of the political context in which the euthanasia debate has been and is occurring. It will also facilitate a more informed debate in the future.

Modelling Cued-Target Recall using Quantum Inspired Models of Target Activation

This article presents and evaluates Quantum Inspired models of Target Activation using Cued-Target Recall Memory Modelling over multiple sources of Free Association data. Two components were evaluated: Whether Quantum Inspired models of Target Activation would provide a better framework than their classical psychological counterparts and how robust these models are across the different sources of Free Association data. In previous work, a formal model of cued-target recall did not exist and as such Target Activation was unable to be assessed directly. Further to that, the data source used was suspected of suffering from temporal and geographical bias. As a consequence, Target Activation was measured against cued-target recall data as an approximation of performance. Since then, a formal model of cued-target recall (PIER3) has been developed [10] with alternative sources of data also becoming available. This allowed us to directly model target activation in cued-target recall with human cued-target recall pairs and use multiply sources of Free Association Data. Featural Characteristics known to be important to Target Activation were measured for each of the data sources to identify any major differences that may explain variations in performance for each of the models. Each of the activation models were used in the PIER3 memory model for each of the data sources and was benchmarked against cued-target recall pairs provided by the University of South Florida (USF). Two methods where used to evaluate performance. The first involved measuring the divergence between the sets of results using the Kullback Leibler (KL) divergence with the second utilizing a previous statistical analysis of the errors [9]. Of the three sources of data, two were sourced from human subjects being the USF Free Association Norms and the University of Leuven (UL) Free Association Networks. The third was sourced from a new method put forward by Galea and Bruza, 2015 in which pseudo Free Association Networks (Corpus Based Association Networks - CANs) are built using co-occurrence statistics on large text corpus. It was found that the Quantum Inspired Models of Target Activation not only outperformed the classical psychological model but was more robust across a variety of data sources.

Student satisfaction with models of field placement supervision

Field placements provide social work students with the opportunity to integrate their classroom learning with the knowledge and skills used in various human service programs. The supervision structure that has most commonly been used is the intensive one-to-one, clinical teaching model. However, this model is being challenged by significant changes in educational and industry sectors, which have led to an increased use of alternative fieldwork structures and supervision arrangements, including task supervision, group supervision, external supervision, and shared supervisory arrangements. This study focuses on identifying models of supervision and student satisfaction with their learning experiences and the supervision received on placement. The study analysed responses to a questionnaire administered to 263 undergraduate social work students enrolled in three different campuses in Australia after they had completed their first or final field placement. The study identified that just over half of the placements used the traditional one student to one social work supervisor model. A number of “emerging” models were also identified, where two or more social workers were involved in the professional supervision of the student. High levels of dissatisfaction were reported by those students who received external social work supervision. Results suggest that students are more satisfied across all aspects of the placement where there is a strong on-site social work presence.