Regularized nonnegative shared subspace learning

Joint modeling of related data sources has the potential to improve various data mining tasks such as transfer learning, multitask clustering, information retrieval etc. However, diversity among various data sources might outweigh the advantages of the joint modeling, and thus may result in performance degradations. To this end, we propose a regularized shared subspace learning framework, which can exploit the mutual strengths of related data sources while being immune to the effects of the variabilities of each source. This is achieved by further imposing a mutual orthogonality constraint on the constituent subspaces which segregates the common patterns from the source specific patterns, and thus, avoids performance degradations. Our approach is rooted in nonnegative matrix factorization and extends it further to enable joint analysis of related data sources. Experiments performed using three real world data sets for both retrieval and clustering applications demonstrate the benefits of regularization and validate the effectiveness of the model. Our proposed solution provides a formal framework appropriate for jointly analyzing related data sources and therefore, it is applicable to a wider context in data mining.

Ridge regression for two dimensional locality preserving projection

Two Dimensional Locality Preserving Projection (2D-LPP) is a recent extension of LPP, a popular face recognition algorithm. It has been shown that 2D-LPP performs better than PCA, 2D-PCA and LPP. However, the computational cost of 2D-LPP is high. This paper proposes a novel algorithm called Ridge Regression for Two Dimensional Locality Preserving Projection (RR- 2DLPP), which is an extension of 2D-LPP with the use of ridge regression. RR-2DLPP is comparable to 2DLPP in performance whilst having a lower computational cost. The experimental results on three benchmark face data sets - the ORL, Yale and FERET databases - demonstrate the effectiveness and efficiency of RR-2DLPP compared with other face recognition algorithms such as PCA, LPP, SR, 2D-PCA and 2D-LPP.

Video genre categorization using audio wavelet coefficients

In this paper, we investigate the use of a wavelet transform-based analysis of audio tracks accompanying videos for the problem of automatic program genre detection. We compare the classification performance based on wavelet-based audio features to that using conventional features derived from Fourier and time analysis for the task of discriminating TV programs such as news, commercials, music shows, concerts, motor racing games, and animated cartoons. Three different classifiers namely the Decision Trees, SVMs, and k-Nearest Neighbours are studied to analyse the reliability of the performance of our wavelet features based approach. Further, we investigate the issue of an appropriate duration of an audio clip to be analyzed for this automatic genre determination. Our experimental results show that features derived from the wavelet transform of the audio signal can very well separate the six video genres studied. It is also found that there is no significant difference in performance with varying audio clip durations across the classifiers.

Information needs and the implications for monitoring health systems : the Australian experience

This paper outlines the information needs underlying the health information system in Australia and the implications these have for the ability to monitor the performance of the health system. We discuss the use of indicators in performance monitoring and the role of information frameworks in providing a basis for their development. The major Australian data sources to support the development of performance indicators are outlined, and their current and likely futures uses for performance monitoring discussed.

Computer simulation based parameter selection for resistance exercise

In contrast to most scientific disciplines, sports science research has been characterized by comparatively little effort investment in the development of relevant phenomenologi-cal models. Scarcer yet is the application of said models in practice. We present a framework which allows resistance training practitioners to employ a recently proposed neu-romuscular model in actual training program design. The first novelty concerns the monitoring aspect of coaching. A method for extracting training performance characteristics from loosely constrained video sequences, effortlessly and with minimal human input, using computer vision is described. The extracted data is subsequently used to fit the underlying neuromuscular model. This is achieved by solving an inverse dynamics problem corresponding to a particular exercise. Lastly, a computer simulation of hypothetical training bouts, using athlete-specific capability parameters, is used to predict the effected adaptation and changes in performance. The software described here allows the practitioner to manipulate hypothetical training parameters and immediately see their effect on predicted adaptation for a specific athlete. Thus, this work presents a holistic view of the monitoring-assessment-adjustment loop.

A manifold approach to face recognition from low quality video across illumination and pose using implicit super-resolution

We consider the problem of matching a face in a low resolution query video sequence against a set of higher quality gallery sequences. This problem is of interest in many applications, such as law enforcement. Our main contribution is an extension of the recently proposed Generic Shape-Illumination Manifold (gSIM) framework. Specifically, (i) we show how super-resolution across pose and scale can be achieved implicitly, by off-line learning of subsampling artefacts; (ii) we use this result to propose an extension to the statistical model of the gSIM by compounding it with a hierarchy of subsampling models at multiple scales; and (iii) we describe an extensive empirical evaluation of the method on over 1300 video sequences – we first measure the degradation in performance of the original gSIM algorithm as query sequence resolution is decreased and then show that the proposed extension produces an error reduction in the mean recognition rate of over 50%.

Conversation pieces

This exhibition presented works by Australian artists Agatha Gothe-Snape, Alex Martinis Roe and Hannah Raisin and collaborative works by Catherine or Kate (Catherine Sagin and Kate Woodcroft), Scott Ferguson (Erika Scott and Brooke Ferguson) and Courtney Coombs and Caitlin Franzmann. These artists engage with ideas of contemporary feminism through processes of dialogue and exchange, exploring subjectivity, humour and intimacy in performance and installation artworks. These works are generated by physical, verbal or textual dialogues between collaborators, participants or the artist and audience; and spark a conversation about feminist art practice today.

Extended conversation pieces

Extended Conversation Pieces brings together conversations about art, and conversation as art. The exhibition features collaborative works by six Brisbane emerging artists, Catherine or Kate (Catherine Sagin and Kate Woodcroft), Scott Ferguson (Erika Scott and Brooke Ferguson) and Courtney Coombs and Caitlin Franzmann. These artists engage with ideas of contemporary feminism through processes of dialogue and exchange; exploring subjectivity, humour and intimacy in performance and installation works. Extended Conversation Pieces showcases the distinctive approach to contemporary practice at Boxcopy, an artist run initiative focused on supporting and commissioning new experimental works by Australian artists, and engaging with collaborative processes and practices.

Restorative care for palliative patients: a retrospective clinical audit of outcomes for patients admitted to an inpatient palliative care unit

BACKGROUND: Restorative care in palliative care is a subset of rehabilitation that aims to improve quality of life through restoration or maintenance of physical functions. Outcomes for restorative care programmes delivered by palliative care units have not adequately been assessed. OBJECTIVES: The objectives are to examine the outcomes of a restorative care programme in an inpatient palliative care unit, including discharge destination, performance status changes and length of stay. METHODS: Retrospective clinical audit of consecutive patients admitted to Calvary Health Care Bethlehem in Melbourne, Australia, principally for restorative care from July 2010 to December 2011. RESULTS: 79 admissions met inclusion criteria. Mean age was 76.5 years (SD=11.14) and 43 (54%) were men. 75 (95%) patients had a malignant diagnosis; of these, the majority had lung cancer (24%). 16 patients (20%) were discharged home, 51 (65%) died and 12 (15%) were transferred. Of the patients discharged home, only 6 (38% of those discharged home) improved their performance status. Those discharged home had a significantly shorter length of stay (17 days compared to 39 days; p<0.05). Patients discharged home also had significantly better Australia-modified Karnofsky Performance Status (AKPS) and Resource Utilisation Groups-Activities of Daily Living (RUG-ADL) scores on admission than others (both p<0.05). CONCLUSIONS: The majority of patients referred for restorative care died during admission, with only a minority discharged home. Patients discharged most commonly experienced maintenance and not improvement in performance status. A successful discharge home following restorative care was associated with a shorter length of stay. Implications and recommendations for successful restorative care will be discussed.

Improved efficiency and stability of flexible dye sensitized solar cells on ITO/PEN substrates using an ionic liquid electrolyte

Flexible dye-sensitized solar cells (DSSCs) built on plastic substrates have attracted great interest as they are lightweight and can be roll-to-roll printed to accelerate production and reduce cost. However, plastic substrates such as PEN and PET are permeable to water, oxygen and volatile electrolyte solvents, which is detrimental to the cell stability. Therefore, to address this problem, in this work, an ionic liquid (IL) electrolyte is used to replace the volatile solvent electrolyte. The initial IL-based devices only achieved around 50% of the photovoltaic conversion efficiency of the cells using the solvent electrolyte. Current-voltage and electrochemical impedance spectroscopy (EIS) analysis of the cells in the dark indicated that this lower efficiency mainly originated from (i) a lack of blocking layer to reduce recombination, and (ii) a lower charge collection efficiency. To combat these problems, cells were developed using a 12 nm thick blocking layer, produced by atomic layer deposition, and 1 μm thick P25 TiO2 film sensitized with the hydrophobic MK-2 dye. These flexible DSSCs utilizing an IL electrolyte exhibit significantly improved efficiencies and a <10% drop in performance after 1000 h aging at 60°C under continuous light illumination.

Enhanced physiological tremor deteriorates plantar flexor torque steadiness after bed rest

This study evaluated the effectiveness of resistance training to preserve submaximal plantar flexor (PF) torque steadiness following 60 days of bed rest (BR). Twenty-two healthy male subjects underwent either BR only (CTR, n=8), or BR plus resistance training (RT, n=14). The magnitude of torque fluctuations during steady submaximal isometric PF contractions (20%, 40%, 60% and 80% of maximum) were assessed before and after BR. Across contraction intensities, torque fluctuations (coefficient of variation, CV) increased more (P<0.05) after BR for CTR (from 0.31±0.10 to 0.92±0.63; P<0.001), than for RT (from 0.30±0.09 to 0.54±0.27; P<0.01). A shift in the spectral content of torque fluctuations towards increased rhythmic activity between 6.5 and 20Hz was observed in CTR only (P<0.05). H-reflex amplitude (H(max)/M(max) ratio) declined across groups from 0.57±0.18 before BR to 0.44±0.14 following BR (P<0.01) without correlation to CV. The present study showed that increased torque fluctuation after BR resulted from enhanced physiological tremor. Resistance training prevented the spectral shift in isometric PF torque fluctuation and offset ∼50% of the decline in performance associated with long-term BR.

Retrospective clinical audit of outcomes for patients admitted to an inpatient palliative care unit for restorative care

Background Restorative care in palliative care is a subset of rehabilitation that aims to improve quality of life through restoration or maintenance of physical functions. Outcomes for restorative care programmes delivered by palliative care units have not adequately been assessed.

Objectives The objectives are to examine the outcomes of a restorative care programme in an inpatient palliative care unit, including discharge destination, performance status changes and length of stay.

Methods Retrospective clinical audit of consecutive patients admitted to Calvary Health Care Bethlehem in Melbourne, Australia, principally for restorative care from July 2010 to December 2011.

Results 79 admissions met inclusion criteria. Mean age was 76.5 years (SD=11.14) and 43 (54%) were men. 75 (95%) patients had a malignant diagnosis; of these, the majority had lung cancer (24%). 16 patients (20%) were discharged home, 51 (65%) died and 12 (15%) were transferred. Of the patients discharged home, only 6 (38% of those discharged home) improved their performance status. Those discharged home had a significantly shorter length of stay (17 days compared to 39 days; p<0.05). Patients discharged home also had significantly better Australia-modified Karnofsky Performance Status (AKPS) and Resource Utilisation Groups-Activities of Daily Living (RUG-ADL) scores on admission than others (both p<0.05).

Conclusions The majority of patients referred for restorative care died during admission, with only a minority discharged home. Patients discharged most commonly experienced maintenance and not improvement in performance status. A successful discharge home following restorative care was associated with a shorter length of stay. Implications and recommendations for successful restorative care will be discussed.

Enhancement of medical named entity recognition using graph-based features

Named Entity Recognition (NER) is a crucial step in text mining. This paper proposes a new graph-based technique for representing unstructured medical text. The new representation is used to extract discriminative features that are able to enhance the NER performance. To evaluate the usefulness of the proposed graph-based technique, the i2b2 medication challenge data set is used. Specifically, the 'treatment' named entities are extracted for evaluation using six different classifiers. The F-measure results of five classifiers are enhanced, with an average improvement of up to 26% in performance.

On-call work: to sleep or not to sleep? It depends

On-call working time arrangements are increasingly common, involve work only in the event of an unpredictable incident and exist primarily outside of standard hours. Like other non-standard working time arrangements, on-call work disrupts sleep and can therefore have negative effects on health, safety and performance. Unlike other non-standard working time arrangements, on-call work often allows sleep opportunities between calls. Any sleep obtained during on-call periods will be beneficial for waking performance. However, there is evidence that sleep while on call may be of substantially reduced restorative value because of the expectation of receiving the call and apprehension about missing the call. In turn, waking from sleep to respond to a call may be associated with temporary increases in performance impairment. This is dependent on characteristics of both the preceding sleep, the tasks required upon waking and the availability and utility of any countermeasures to support the transition from sleep to wake. In this paper, we critically evaluate the evidence both for and against sleeping during on-call periods and conclude that some sleep, even if it is of reduced quality and broken by repeated calls, is a good strategy. We also note, however, that organisations utilising on-call working time arrangements need to systematically manage the likelihood that on-call sleep can be associated with temporary performance impairments upon waking. Given that the majority of work in this area has been laboratory-based, there is a significant need for field-based investigations of the magnitude of sleep inertia, in addition to the utility of sleep inertia countermeasures. Field studies should include working with subject matter experts to identify the real-world impacts of changes in performance associated with sleeping, or not sleeping, whilst on call.

Validation of jump squats as a practical measure of post-activation potentiation

To determine if post-activation potentiation (PAP) can augment sports performance, it is pertinent that researchers be confident that any enhancement in performance is attributable to the PAP phenomenon. However, obtaining mechanistic measures of PAP in the daily training environment of highly trained athletes is impractical. We sought to validate jump squats as a practical measure with ecological validity to sports performance against a mechanistic measure of PAP. We assessed the evoked muscle twitch properties of the knee extensors and jump squat kinetics of 8 physically trained males in response to a 5-repetition-maximum back squat conditioning stimulus (CS). Evoked muscle twitch, followed by 3 jump squats, was assessed before and at 4, 8, and 12 min post CS. Time intervals were assessed on separate occasions using a Latin square design. Linear regression was used to determine the relationship between post-pre changes in kinetic variables and muscle twitch peak force (Ft) and twitch rate of force development (RFDt). Large correlations were observed for both concentric relative and absolute mean power and Ft (r = 0.50 ± 0.30) and RFDt (r = 0.56 ± 0.27 and r = 0.58 ± 0.26). Concentric rate of force development (RFD) showed moderate correlations with Ft (r = 0.45 ± 0.33) and RFDt (r = 0.49 ± 0.32). Small-to-moderate correlations were observed for a number of kinetic variables (r = -0.42-0.43 ± 0.32-0.38). Jump squat concentric mean power and RFD are valid ecological measures of muscle potentiation, capable of detecting changes in athletic performance in response to the PAP phenomenon.