An integrated framework for suicide risk prediction

Suicide is a major concern in society. Despite of great attention paid by the community with very substantive medico-legal implications, there has been no satisfying method that can reliably predict the future attempted or completed suicide. We present an integrated machine learning framework to tackle this challenge. Our proposed framework consists of a novel feature extraction scheme, an embedded feature selection process, a set of risk classifiers and finally, a risk calibration procedure. For temporal feature extraction, we cast the patient’s clinical history into a temporal image to which a bank of one-side filters are applied. The responses are then partly transformed into mid-level features and then selected in 1-norm framework under the extreme value theory. A set of probabilistic ordinal risk classifiers are then applied to compute the risk probabilities and further re-rank the features. Finally, the predicted risks are calibrated. Together with our Australian partner, we perform comprehensive study on data collected for the mental health cohort, and the experiments validate that our proposed framework outperforms risk assessment instruments by medical practitioners.

Evidencing learning outcomes: a multi-level, multi-dimensional course alignment model

This conceptual framework proposes a multi-level, multi-dimensional course alignment model to implement a contextualised constructive alignment of rubric design that authentically evidences and assesses learning outcomes. By embedding quality control mechanisms at each level for each dimension, this model facilitates the development of an aligned curriculum. The use of rubrics is a crucial theme for many higher education institutions owing to the binding requirement by universities to provide evidence to quality assurance agencies. The success of evidencing learning outcomes through rubrics, however, is only one piece of the puzzle. The other is the contextualised constructive alignment of intertwined factors. Despite the significance of embedding these factors, there has been little, if any, systematic framework in this area. The two key instrumental forces underpinning the conception of this model are: seeking external accreditation and the implementation of programme enhancement thus realising the strategic agenda for an Australian university.

The ‘r’ word and closure of remote Aboriginal communities in Western Australia; a potential framework for change

AbstractThe latest Australian Commonwealth Government Close the Gap Report reveals the circumstances of many of Australia’s Indigenous Peoples are either stagnant or going backwards. This paper argues that such ongoing injustice is a consequence of systemic racism that has been perpetuated since colonization and sustained in the twenty first century by discussion or mention of racism being taboo. A counter colonial educational framework is then provided that has the potential to address such institutional racism. The paper begins by providing a definition of systemic racism. Following this there is a brief explanation of the unique geographical context and the racist history of colonization in Australia. The nature of remote communities, the link between traditional law, country and identity will be outlined. Based on readily available sources such as media reports, social media links, and public policy announcements by government the paper then reflects on what has been reported about closure of remote communities in Western Australia. Government policy, announcements and events of the past year will be described and critically discussed in light of the definition of racism provided at the beginning of the article. The proposed framework requires self-reflexivity of organisations and individuals with a particular focus on aspects of sovereignty, healing, re-learning history and starting with a focus on agency instead of deficit. Being guided by this framework has the potential to avoid arbitrarily forcing people from their physical, spiritual and ancestral home, though this is likely to be a long term proposition rather than a quick fix.

Structured representation learning from complex data

This thesis advances several theoretical and practical aspects of the recently introduced restricted Boltzmann machine - a powerful probabilistic and generative framework for modelling data and learning representations. The contributions of this study represent a systematic and common theme in learning structured representations from complex data.

Fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images via a learning-based method

In this paper, we address the problems of fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images. We propose a learning-based, unified random forest regression and classification framework to tackle these two problems. More specifically, in the first stage, the localization of 3D vertebral bodies is solved with random forest regression where we aggregate the votes from a set of randomly sampled image patches to get a probability map of the center of a target vertebral body in a given image. The resultant probability map is then further regularized by Hidden Markov Model (HMM) to eliminate potential ambiguity caused by the neighboring vertebral bodies. The output from the first stage allows us to define a region of interest (ROI) for the segmentation step, where we use random forest classification to estimate the likelihood of a voxel in the ROI being foreground or background. The estimated likelihood is combined with the prior probability, which is learned from a set of training data, to get the posterior probability of the voxel. The segmentation of the target vertebral body is then done by a binary thresholding of the estimated probability. We evaluated the present approach on two openly available datasets: 1) 3D T2-weighted spine MR images from 23 patients and 2) 3D spine CT images from 10 patients. Taking manual segmentation as the ground truth (each MR image contains at least 7 vertebral bodies from T11 to L5 and each CT image contains 5 vertebral bodies from L1 to L5), we evaluated the present approach with leave-one-out experiments. Specifically, for the T2-weighted MR images, we achieved for localization a mean error of 1.6 mm, and for segmentation a mean Dice metric of 88.7% and a mean surface distance of 1.5 mm, respectively. For the CT images we achieved for localization a mean error of 1.9 mm, and for segmentation a mean Dice metric of 91.0% and a mean surface distance of 0.9 mm, respectively.

Lessons in building capacity in sexuality education using the health promoting school framework: from planning to implementation

Purpose – The health promoting school model is rarely implemented in relation to sexuality education. This paper reports on data collected as part of a five-year project designed to implement a health promoting and whole school approach to sexuality education in a five campus year 1-12 college in regional Victoria, Australia. Using a community engagement focus involving local and regional stakeholders and with a strong research into practice component, the project is primarily concerned with questions of capacity building, impact and sustainability as part of whole school change. The paper aims to discuss this issue. Design/methodology/approach – Using an action research design, data were collected from parents, students, teachers and key community stakeholders using a mixed methods approach involving surveys, interviews, document analysis and participant observation. Findings – Sexuality education has become a key school policy and has been implemented from years 1 to 9. Teachers and key support staff have engaged in professional learning, a mentor program has been set up, a community engagement/parent liaison position has been created, and parent forums have been conducted on all five campuses. Research limitations/implications – The translation of research into practice can be judged by the impact it has on teacher capacity and the students’ experience. Classroom observation and more longitudinal research would shed light on whether the espoused changes are happening in reality. Originality/value – This paper reports on lessons learned and the key enabling factors that have built capacity to ensure that sexuality education within a health promoting, whole school approach will remain sustainable into the future. These findings will be relevant to others interested in building capacity in sexuality education and health promotion more generally.

Learning clinical skills during bedside teaching encounters in general practice: a video-observational study with insights from activity theory

Purpose – This paper aims to explore how opportunities for learning clinical skills are negotiated within bedside teaching encounters (BTEs). Bedside teaching, within the medical workplace, is considered essential for helping students develop their clinical skills. Design/methodology/approach – An audio and/or video observational study examining seven general practice BTEs was undertaken. Additionally, audio-recorded, semi-structured interviews were conducted with participants. All data were transcribed. Data analysis comprised Framework Analysis informed by Engeström’s Cultural Historical Activity Theory. Findings – BTEs can be seen to offer many learning opportunities for clinical skills. Learning opportunities are negotiated by the participants in each BTE, with patients, doctors and students playing different roles within and across the BTEs. Tensions emerged within and between nodes and across two activity systems. Research limitations/implications – Negotiation of clinical skills learning opportunities involved shifts in the use of artefacts, roles and rules of participation, which were tacit, dynamic and changing. That learning is constituted in the activity implies that students and teachers cannot be fully prepared for BTEs due to their emergent properties. Engaging doctors, students and patients in refecting on tensions experienced and the factors that infuence judgements in BTEs may be a useful frst step in helping them better manage the roles and responsibilities therein. Originality/value – The paper makes an original contribution to the literature by highlighting the tensions inherent in BTEs and how the negotiation of roles and division of labour whilst juggling two interacting activity systems create or inhibit opportunities for clinical skills learning. This has signifcant implications for how BTEs are conceptualised.

A predictive framework for modeling healthcare data with evolving clinical interventions

Medical interventions critically determine clinical outcomes. But prediction models either ignore interventions or dilute impact by building a single prediction rule by amalgamating interventions with other features. One rule across all interventions may not capture differential effects. Also, interventions change with time as innovations are made, requiring prediction models to evolve over time. To address these gaps, we propose a prediction framework that explicitly models interventions by extracting a set of latent intervention groups through a Hierarchical Dirichlet Process (HDP) mixture. Data are split in temporal windows and for each window, a separate distribution over the intervention groups is learnt. This ensures that the model evolves with changing interventions. The outcome is modeled as conditional, on both the latent grouping and the patients' condition, through a Bayesian logistic regression. Learning distributions for each time-window result in an over-complex model when interventions do not change in every time-window. We show that by replacing HDP with a dynamic HDP prior, a more compact set of distributions can be learnt. Experiments performed on two hospital datasets demonstrate the superiority of our framework over many existing clinical and traditional prediction frameworks.

‘I’m not like that, why treat me the same way?’ The impact of stereotyping international students on their learning, employability and connectedness with the workplace

A significant body of literature on international education examines the experiences of international students in the host country. There is however a critical lack of empirical work that investigates the dynamic and complex positioning of international students within the current education-migration nexus that prevails international education in countries such as Australia, Canada and the UK. This paper addresses an important but under-researched area of the education-migration landscape by examining how the stereotyping of students as mere ‘migration hunters’ may impact their study and work experiences. It draws on a four-year research project funded by the Australian Research Council that includes more than 150 interviews and fieldwork in the Australian vocational education context. Positioning theory is used as a conceptual framework to analyse how generalising international students as ‘mere migration hunters’ has led to the disconnectedness, vulnerability and marginalization of the group of international students participating in this research.

What are the perceived learning needs of Australian general practice registrars for quality prescribing?

BACKGROUND: Little is known about the perceived learning needs of Australian general practice (GP) registrars in relation to the quality use of medicines (QUM) or the difficulties experienced when learning to prescribe. This study aimed to address this gap. METHODS: GP registrars' perceived learning needs were investigated through an online national survey, interviews and focus groups. Medical educators' perceptions were canvassed in semi-structured interviews in order to gain a broader perspective of the registrars' needs. Qualitative data analysis was informed by a systematic framework method involving a number of stages. Survey data were analysed descriptively. RESULTS: The two most commonly attended QUM educational activities took place in the workplace and through regional training providers. Outside of these structured educational activities, registrars learned to prescribe mainly through social and situated means. Difficulties encountered by GP registrars included the transition from hospital prescribing to prescribing in the GP context, judging how well they were prescribing and identifying appropriate and efficient sources of information at the point of care. CONCLUSIONS: GP registrars learn to prescribe primarily and opportunistically in the workplace. Despite many resources being expended on the provision of guidelines, decision-support systems and training, GP registrars expressed difficulties related to QUM. Ways of easing the transition into GP and of managing the information 'overload' related to medicines (and prescribing) in an evidence-guided, efficient and timely manner are needed. GP registrars should be provided with explicit feedback about the process and outcomes of prescribing decisions, including the use of audits, in order to improve their ability to judge their own prescribing.

Detection of dynamic background due to swaying movements from motion features

Dynamically changing background (dynamic background) still presents a great challenge to many motion-based video surveillance systems. In the context of event detection, it is a major source of false alarms. There is a strong need from the security industry either to detect and suppress these false alarms, or dampen the effects of background changes, so as to increase the sensitivity to meaningful events of interest. In this paper, we restrict our focus to one of the most common causes of dynamic background changes: 1) that of swaying tree branches and 2) their shadows under windy conditions. Considering the ultimate goal in a video analytics pipeline, we formulate a new dynamic background detection problem as a signal processing alternative to the previously described but unreliable computer vision-based approaches. Within this new framework, we directly reduce the number of false alarms by testing if the detected events are due to characteristic background motions. In addition, we introduce a new data set suitable for the evaluation of dynamic background detection. It consists of real-world events detected by a commercial surveillance system from two static surveillance cameras. The research question we address is whether dynamic background can be detected reliably and efficiently using simple motion features and in the presence of similar but meaningful events, such as loitering. Inspired by the tree aerodynamics theory, we propose a novel method named local variation persistence (LVP), that captures the key characteristics of swaying motions. The method is posed as a convex optimization problem, whose variable is the local variation. We derive a computationally efficient algorithm for solving the optimization problem, the solution of which is then used to form a powerful detection statistic. On our newly collected data set, we demonstrate that the proposed LVP achieves excellent detection results and outperforms the best alternative adapted from existing art in the dynamic background literature.

Project-oriented design-based learning in engineering education

The School of Engineering at Deakin University (SOE-DU) is committed to provide authentic and unique learning experiences to students. Over the last five years, SOE-DU has undertaken a study to develop a unique teaching and learning model. The proposed model was based on the Project- Oriented Design Based Learning (PODBL) philosophy which is unique within Australia and in the world. Fundamentally, the framework balances project-driven pedagogy with a design-focused practice in response to industry needs. This paper focuses on the development of PODBL and articulating how it helps nurture creative and industry-ready professional engineers.

Multiple task transfer learning with small sample sizes

Prognosis, such as predicting mortality, is common in medicine. When confronted with small numbers of samples, as in rare medical conditions, the task is challenging. We propose a framework for classification with data with small numbers of samples. Conceptually, our solution is a hybrid of multi-task and transfer learning, employing data samples from source tasks as in transfer learning, but considering all tasks together as in multi-task learning. Each task is modelled jointly with other related tasks by directly augmenting the data from other tasks. The degree of augmentation depends on the task relatedness and is estimated directly from the data. We apply the model on three diverse real-world data sets (healthcare data, handwritten digit data and face data) and show that our method outperforms several state-of-the-art multi-task learning baselines. We extend the model for online multi-task learning where the model parameters are incrementally updated given new data or new tasks. The novelty of our method lies in offering a hybrid multi-task/transfer learning model to exploit sharing across tasks at the data-level and joint parameter learning.

A framework for classifying online mental health related communities with an interest in depression

Mental illness has a deep impact on individuals, families, and by extension, society as a whole. Social networks allow individuals with mental disorders to communicate with others sufferers via online communities, providing an invaluable resource for studies on textual signs of psychological health problems. Mental disorders often occur in combinations, e.g., a patient with an anxiety disorder may also develop depression. This co-occurring mental health condition provides the focus for our work on classifying online communities with an interest in depression. For this, we have crawled a large body of 620,000 posts made by 80,000 users in 247 online communities. We have extracted the topics and psycho-linguistic features expressed in the posts, using these as inputs to our model. Following a machine learning technique, we have formulated a joint modelling framework in order to classify mental health-related co-occurring online communities from these features. Finally, we performed empirical validation of the model on the crawled dataset where our model outperforms recent state-of-the-art baselines.

Learning feature trajectories using Gabor Filter Bank for human activity segmentation and recognition

We describe a novel method for human activity segmentation and interpretation in surveillance applications based on Gabor filter-bank features. A complex human activity is modeled as a sequence of elementary human actions like walking, running, jogging, boxing, hand-waving etc. Since human silhouette can be modeled by a set of rectangles, the elementary human actions can be modeled as a sequence of a set of rectangles with different orientations and scales. The activity segmentation is based on Gabor filter-bank features and normalized spectral clustering. The feature trajectories of an action category are learnt from training example videos using Dynamic Time Warping. The combined segmentation and the recognition processes are very efficient as both the algorithms share the same framework and Gabor features computed for the former can be used for the later. We have also proposed a simple shadow detection technique to extract good silhouette which is necessary for good accuracy of an action recognition technique. © 2008 IEEE.