A framework for mixed-type multi-outcome prediction with applications in healthcare

Health analysis often involves prediction of multiple outcomes of mixed-type. Existing work is restrictive to either a limited number or specific outcome types. We propose a framework for mixed-type multi-outcome prediction. Our proposed framework proposes a cumulative loss function composed of a specific loss function for each outcome type - as an example, least square (continuous outcome), hinge (binary outcome), poisson (count outcome) and exponential (non-negative outcome). Tomodel these outcomes jointly, we impose a commonality across the prediction parameters through a common matrix-Normal prior. The framework is formulated as iterative optimization problems and solved using an efficient Block coordinate descent method (BCD). We empirically demonstrate both scalability and convergence. We apply the proposed model to a synthetic dataset and then on two real-world cohorts: a Cancer cohort and an Acute Myocardial Infarction cohort collected over a two year period. We predict multiple emergency related outcomes - as example, future emergency presentations (binary), emergency admissions (count), emergency length-of-stay-days (non-negative) and emergency time-to-next-admission-day (non-negative). Weshow that the predictive performance of the proposed model is better than several state-of-the-art baselines.

Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM)

Electronic medical record (EMR) offers promises for novel analytics. However, manual feature engineering from EMR is labor intensive because EMR is complex - it contains temporal, mixed-type and multimodal data packed in irregular episodes. We present a computational framework to harness EMR with minimal human supervision via restricted Boltzmann machine (RBM). The framework derives a new representation of medical objects by embedding them in a low-dimensional vector space. This new representation facilitates algebraic and statistical manipulations such as projection onto 2D plane (thereby offering intuitive visualization), object grouping (hence enabling automated phenotyping), and risk stratification. To enhance model interpretability, we introduced two constraints into model parameters: (a) nonnegative coefficients, and (b) structural smoothness. These result in a novel model called eNRBM (EMR-driven nonnegative RBM). We demonstrate the capability of the eNRBM on a cohort of 7578 mental health patients under suicide risk assessment. The derived representation not only shows clinically meaningful feature grouping but also facilitates short-term risk stratification. The F-scores, 0.21 for moderate-risk and 0.36 for high-risk, are significantly higher than those obtained by clinicians and competitive with the results obtained by support vector machines.

Sexual and violent recidivism by offender type and actuarial risk : reoffending rates for rapists, child molesters and mixed-victim offenders

The current study examines sexual and violent reoffence rates for a sample of 2474 sexual offenders over an average of 15 years following release from prison. Reoffence rates are reported as a function of the offenders' victim type and level of risk as assessed by the Automated Sexual Recidivism Scale, a computer scored measure of relevant historical risk factors. Observed sexual recidivism rates for offenders with child victims, adult victims, and mixed victims were quite similar. Results indicate that offenders with exclusively female child victims not only showed a lower rate of sexual reoffending, but that the reoffence rates were relatively low across all levels of actuarial risk. In contrast, those with male child victims and adult victims showed a pronounced escalation of reoffence rates as actuarial risk increased. Results also indicated that adult victim offenders are less consistent in the victim type of their reoffences, with 37% sexually reoffending against child victims. Finally, combined rates of sexual and violent reoffending were particularly high for those with adult victim sexual offence histories. Risk assessment and public policy implications are discussed.

Research progress and materials selection guidelines on mixed conducting perovskite-type ceramic membranes for oxygen production

Oxygen production by air separation is of great importance in both environmental and industrial processes as most large scale clean energy technologies require oxygen as feed gas. Currently the conventional cryogenic air separation unit is a major economic impediment to the deployment of these clean energy technologies with carbon capture (i.e. oxy-fuel combustion ). Dense ceramic perovskite membranes are envisaged to replace the cryogenics and reduce O2 production costs by 35% or more; which can significantly cut the energy penalty by 50% when integrated in oxy-fuel power plant for CO2 capture. This paper reviews the current progress in the development of dense ceramic membranes for oxygen production. The principles, advantages or disadvantages, and the crucial problems of all kinds of membranes are discussed. Materials development, optimisation guidelines and suggestions for future research direction are also included. Some areas already previously reviewed are treated with less attention.

Managing design knowledge with mixed-initiative dialogue

This paper is based on ongoing work in developing interactive interfaces to formal methods for encoding design knowledge. It reports on the development of a shared graphical notation to support user interaction with design knowledge based on mixed-initiative. Mixed-initiative provides a model of interaction where both the designer and the knowledge formalism may share responsibility over decisions. The paper discusses how a formal visual notation can support the mixed-initiative mode for developing and managing formal design knowledge. The notation addresses on the dialogue problem between the user and a knowledge based formalism and illustrates a model of interaction in which the user and the formalism can share and input data through a common shared resource, on a common shared task. The paper demonstrates the use of this notation in common decision tasks and the implications for seamless interaction with design support systems.
type="hidden" id="gwProxy" />type="hidden" onclick="jsCall();" id="jsProxy" />

Expression of natriuretic peptide receptor mRNA and functional response to atrial natriuretic peptide and C-type natriuretic peptide in rainbow trout (Oncorhynchus mykiss) head kidney leucocytes

The stimulatory effect of vasomodulatory natriuretic peptide hormones on macrophages and peripheral blood leucocytes in mammals is well-established. However, the relationship in lower vertebrates has not been characterised. Expression of atrial natriuretic peptide, ventricular natriuretic peptide and C-type natriuretic peptide-1, and the guanylyl cyclase-linked (GC) natriuretic peptide receptor-A and -B-type receptors (NPR-A and NPR-B, respectively) was determined by PCR from the mRNA of rainbow trout head kidney leucocytes yielding gene fragments with 100% homology to the same respective natriuretic peptide and NPR-A and -B sequences obtained from other rainbow trout tissues. A mixed population of isolated rainbow trout head kidney leucocytes was stimulated in vitro with trout atrial natriuretic peptide (specific NPR-A agonist) and trout C-type natriuretic peptide (NPR-A and -B agonist) as well as the cGMP agonist 8-bromo-cGMP or the GC inhibitor 8-bromo-phenyl-eutheno-cGMP. Respiratory burst was stimulated by trout atrial natriuretic peptide, trout C-type natriuretic peptide-1 and 8-bromo-cGMP in a dose dependant manner with the highest activity as a result of stimulation with trout C-type natriuretic peptide-1 in excess of that achieved by phorbol myristate acetate (PMA). Equimolar concentrations of the inhibitor, inhibited the respiratory burst caused by the natriuretic peptides and 8-bromo-cGMP. The natriuretic peptide receptors on rainbow trout head kidney leucocytes appear to have a stimulatory function with regard to respiratory burst that is activated through a cGMP second messenger pathway and the natriuretic peptides expressed in the head kidney leucocytes may well act in a paracrine/autocrine manner.

Influence of gender and pair-type on preschool children's learning in computing context

The thesis examined early differences between girls and boys in their attitudes and social behaviour which might help explain why girls and women continue to reject computing. The behaviour of preschool children playing freely for three supervised sessions in either same-gender or mixed-gender pairs with a computerised robot was studied, and then their individual programming performance was measured. Conclusions were that social interaction and computer programming performance were not differentiated by gender. Mixed-gender pairs had a significant effect on both children's style of social interaction and their performance, an impact that was particularly negative for girls.

Mixed transfer function neural networks for knowledge acquistition

Modeling helps to understand and predict the outcome of complex systems. Inductive modeling methodologies are beneficial for modeling the systems where the uncertainties involved in the system do not permit to obtain an accurate physical model. However inductive models, like artificial neural networks (ANNs), may suffer from a few drawbacks involving over-fitting and the difficulty to easily understand the model itself. This can result in user reluctance to accept the model or even complete rejection of the modeling results. Thus, it becomes highly desirable to make such inductive models more comprehensible and to automatically determine the model complexity to avoid over-fitting. In this paper, we propose a novel type of ANN, a mixed transfer function artificial neural network (MTFANN), which aims to improve the complexity fitting and comprehensibility of the most popular type of ANN (MLP - a Multilayer Perceptron).

A prospective study of the impact of subthreshold mixed states on the 24-month clinical outcomes of bipolar I disorder or schizoaffective disorder

Objectives: Subthreshold mixed states are common, yet their clinical significance is unclear. This study investigated the clinical outcomes of subthreshold mixed states in participants with bipolarI disorder or schizoaffective disorder, using the Cassidy and Benazzi criteria for manic and depressive mixed states, respectively.

Methods: The Bipolar Comprehensive Outcomes Study (BCOS) is a prospective observational study of treatment and outcomes for patients with bipolar I or schizoaffective disorder, bipolar type. Participants (N=239) were grouped based on study entry clinical presentation as having pure depression (n=63) if they satisfied DSM-IV-TR criteria for a Major Depressive Episode (MDE), depressive mixed state (DMX) if they also had at least three concurrent hypomanic symptoms (n=33), or not depressed (n=143) if they did not satisfy the criteria for MDE. Participants were similarly grouped as having pure mania (n=3) if they satisfied DSM-IV criteria for a Manic Episode, manic mixed state (MMX) if they also had at least two concurrent depressive symptoms (n=33), or not manic (n=203). Clinical data were collected by interview every 3-months over a 24-month period.

Results: Measures of quality of life, mental and physical health over the 24-month period were significantly worse for participants who were classified as having mixed states at study entry on most outcome measures compared to participants who were not in an illness episode at study entry. DMX was predictive of greater manic symptomatology over the 24 months compared to participants with pure depression.

Conclusion: In participants with a current episode of mood disorder, the presence of subthreshold symptoms of opposite polarity was associated with poorer clinical outcomes over a 24-month period.

Patient-reported outcomes following islet cell or pancreas transplantation (alone or after kidney) in type 1 diabetes: a systematic review

Aims For selected individuals with complex Type 1 diabetes, pancreatic islet transplantation (IT) offers the potential of excellent glycaemic controlwithout significant hypoglycaemia, balanced by the need for ongoing systemic immunosuppression. Increasingly, patient-reported outcomes (PROs) are considered alongside biomedical outcomes as a measure of transplant success. PROs in IT have not previously been compared directlywith the closest alternate treatment option, pancreas transplant alone (PTA) or pancreas after kidney (PAK).

Methods We used a Population, Intervention, Comparisons, Outcomes (PICO) strategy to search Scopus and screened 314 references for inclusion.

Results Twelve studies [including PRO assessment of PAK, PTA, islet-after kidney (IAK) and islet transplant alone (ITA); n = 7–205] used a total of nine specified and two unspecified PRO measures. Results were mixed but identified some benefits which remained apparent up to 36 months post-transplant, including improvements in fear of hypoglycaemia, as well as some aspects of diabetes-specific quality of life (QoL) and general health status. Negative outcomes included short-term pain associated with the procedure, immunosuppressant side effects and depressed mood associated with loss of graft function.

Conclusions The mixed resultsmay be attributable to limited sample sizes. Also, some PROmeasures may lack sensitivity to detect actual changes, as they exclude issues and domains of life likely to be important forQoL post-transplantation and when patients may no longer perceive themselves to have diabetes. Thus, the full impact of islet ⁄ pancreas transplantation (alone or after kidney) on QoL is unknown. Furthermore, no studies have assessed patient satisfaction, which may highlight further advantages and disadvantages of transplantation.

Does self monitoring of blood blucose as opposed to urinalysis provide additional benefit in patients newly diagnosed with type 2 diabetes receiving structured education? The DESMOND SMBG randomised controlled trial protocol

Background
The benefit of self-monitoring of blood glucose (SMBG) in people with type 2 diabetes on diet or oral agents other than sulphonylureas remains uncertain. Trials of interventions incorporating education about self-monitoring of blood glucose have reported mixed results. A recent systematic review concluded that SMBG was not cost-effective. However, what was unclear was whether a cheaper method of self-monitoring (such as urine glucose monitoring) could produce comparable benefit and patient acceptability for less cost.

Methods/Design
The DESMOND SMBG trial is comparing two monitoring strategies (blood glucose monitoring and urine testing) over 18 months when incorporated into a comprehensive self-management structured education programme. It is a multi-site cluster randomised controlled trial, conducted across 8 sites (7 primary care trusts) in England, UK involving individuals with newly diagnosed Type 2 diabetes.

The trial has 80% power to demonstrate equivalence in mean HbA1c (the primary end-point) at 18 months of within ± 0.5% assuming 20% drop out and 20% non-consent. Secondary end-points include blood pressure, lipids, body weight and psychosocial measures as well as a qualitative sub-study.

Practices were randomised to one of two arms: participants attend a DESMOND programme incorporating a module on self-monitoring of either urine or blood glucose. The programme is delivered by accredited educators who received specific training about equipoise. Biomedical data are collected and psychosocial scales completed at baseline, and 6, 12, and 18 months post programme. Qualitative research with participants and educators will explore views and experiences of the trial and preferences for methods of monitoring.

Discussion
The DESMOND SMBG trial is designed to provide evidence to inform the debate about the value of self-monitoring of blood glucose in people with newly diagnosed type 2 diabetes. Strengths include a setting in primary care, a cluster design, a health economic analysis, a comparison of different methods of monitoring while controlling for other components of training within the context of a quality assured structured education programme and a qualitative sub-study.

Impacts of mixed-use development on parking requirements

The objective of this study is to investigate the impacts of mixed use developments on parking requirements. Along with the renewed interests in mixed use development, shared parking concept has been a focus for traffic engineers, local governments and mixed use developers in recent years. With the help of a case study of Gold Coast city in Australia, this study has identified that shared parking significantly reduces the overall parking requirements of mixed use developments. However, to be shared parking more effective, the type and the size of various land uses within a particular mixed use development should favour the concept. For example, offices and hotels can go side by side as the time-of-day parking requirements and peak parking demand do not conflict each other. A series of time-of-day parking occupancy rates have been developed for typical land use categories to identify such effectiveness.

Investigating retention characteristics of a mixed-mode stationary phase and the enhancement of monolith selectivity for high-performance liquid chromatography

The synthesis and chromatographic behavior of an analytical size mixed-mode bonded silica monolith was investigated. The monolith was functionalized by an in situ modification process of a bare silica rod with chloro(3-cyanopropyl)dimethyl silane and chlorodimethyl propyl phenyl silane solutions. These ligands were selected in order to combine both resonance and nonresonance π-type bonding within a single separation environment. Selectivity studies were undertaken using n-alkyl benzenes and polycyclic aromatic hydrocarbons in aqueous methanol and acetonitrile mobile phases to assess the methylene and aromatic selectivities of the column. The results fit with the linear solvent strength theory suggesting excellent selectivity of the column was achieved. Comparison studies were performed on monolithic columns that were functionalized separately with cyano and phenyl ligands, suggesting highly conjugated molecules were able to successfully exploit both of the π-type selectivities afforded by the two different ligands on the mixed-mode column.

Sociocultural factors and perceptions associated with type 2 diabetes among sub-Saharan African migrants in Melbourne, Victoria

Aim: To explore how non-diabetic sub-Saharan African migrants residing in Melbourne, construct and interpret type 2 diabetes mellitus (T2DM) and its risk factors and to provide an evidence-based theoretical framework to inform community-based prevention programs. Methods: Seven focus group discussions (two with women only, two with men only and three of mixed gender) were carried out among Ghanaian, Zimbabwean, Sudanese and Burundian migrants living in Melbourne (n = 61; age range: 18 to 61 years). Results: Three distinct themes emerged: not paying much attention to the threat of T2DM and othering; T2DM being outside the individuals' control; and entrapment within rich industrialised culture and lifestyle. Participants perceived T2DM to be a disease of the wealthy caused by intake of too much sugar and sedentary behaviour, which were particularly compounded by lifestyle-related changes upon migration to an industrialised country. However, they also perceived T2DM to be associated with bad luck. Conclusions: Culturally competent prevention and education programs are needed to increase health literacy and dispel religious and cultural myths about T2DM among sub-Saharan African migrants.

First report of a phytogeographically mixed (transitional) Middle-Late Permian fossil wood assemblage from the Hami area, northwest China, and implications for Permian phytogeographical, paleogeographical and paleoclimatic evolution in central Asia

A diverse and well-preserved fossil wood assemblage is described, for the first time, from the Middle Permian Taerlang Formation and the Upper Permian Quanzijie Formation in the vicinity of the Tianshan Town, Hami City of northwestern China. On the basis of wood microstructure, the fossil woods are classified into three genera and five species, including one new genus: Prototianshanoxylon gen. nov. and two new species: Prototianshanoxylon erdaogouense sp. nov., Prototianshanoxylon hamiense sp. nov. The new genus is characterized by window-like cross-field pitting and mixed tracheid radial wall pitting that suggest a transitional type between araucarioid-type and protopinoid-type pittings.Phytogeographically, the fossil wood assemblage is characterized by an admixture of elements of both temperate Angaran (represented by wood specimens with moderately to well defined growth rings in their secondary xylem) and tropical-subtropical north subregion of the Cathaysian floras (with wood specimens lacking well-defined growth rings). Such a phytogeographically mixed fossil wood assemblage is interpreted to represent a transitional and complex climate condition between a cool temperate and tropical to subtropical zones, showing both seasonal variation and unstable climate conditions. Previously, similarly mixed floras have already been found to exist widely in northern China ranging in age from Early to Late Permian, but the mechanisms thought to be responsible for their formation were varied and remain controversial. In this study, the formation of these mixed Permian floras of North China is linked to the closure of the Tianshan-Hingan seaway coupled with the collision and amalgamation of Siberia with North China and the Tarim block, in a manner much like closing a pair of scissors with the closure of the seaway proceeding gradually and progressively from west to the east.