Injury narrative text classification : a preliminary study

Description of a patient's injuries is recorded in narrative text form by hospital emergency departments. For statistical reporting, this text data needs to be mapped to pre-defined codes. Existing research in this field uses the Naïve Bayes probabilistic method to build classifiers for mapping. In this paper, we focus on providing guidance on the selection of a classification method. We build a number of classifiers belonging to different classification families such as decision tree, probabilistic, neural networks, and instance-based, ensemble-based and kernel-based linear classifiers. An extensive pre-processing is carried out to ensure the quality of data and, in hence, the quality classification outcome. The records with a null entry in injury description are removed. The misspelling correction process is carried out by finding and replacing the misspelt word with a soundlike word. Meaningful phrases have been identified and kept, instead of removing the part of phrase as a stop word. The abbreviations appearing in many forms of entry are manually identified and only one form of abbreviations is used. Clustering is utilised to discriminate between non-frequent and frequent terms. This process reduced the number of text features dramatically from about 28,000 to 5000. The medical narrative text injury dataset, under consideration, is composed of many short documents. The data can be characterized as high-dimensional and sparse, i.e., few features are irrelevant but features are correlated with one another. Therefore, Matrix factorization techniques such as Singular Value Decomposition (SVD) and Non Negative Matrix Factorization (NNMF) have been used to map the processed feature space to a lower-dimensional feature space. Classifiers with these reduced feature space have been built. In experiments, a set of tests are conducted to reflect which classification method is best for the medical text classification. The Non Negative Matrix Factorization with Support Vector Machine method can achieve 93% precision which is higher than all the tested traditional classifiers. We also found that TF/IDF weighting which works well for long text classification is inferior to binary weighting in short document classification. Another finding is that the Top-n terms should be removed in consultation with medical experts, as it affects the classification performance.

What patient information allows us to make accurate predictions of outcome?

Only some of the information contained in a medical record will be useful to the prediction of patient outcome. We describe a novel method for selecting those outcome predictors which allow us to reliably discriminate between adverse and benign end results. Using the area under the receiver operating characteristic as a nonparametric measure of discrimination, we show how to calculate the maximum discrimination attainable with a given set of discrete valued features. This upper limit forms the basis of our feature selection algorithm. We use the algorithm to select features (from maternity records) relevant to the prediction of failure to progress in labour. The results of this analysis motivate investigation of those predictors of failure to progress relevant to parous and nulliparous sub-populations.

Feature selection using expected attainable discrimination

We propose expected attainable discrimination (EAD) as a measure to select discrete valued features for reliable discrimination between two classes of data. EAD is an average of the area under the ROC curves obtained when a simple histogram probability density model is trained and tested on many random partitions of a data set. EAD can be incorporated into various stepwise search methods to determine promising subsets of features, particularly when misclassification costs are difficult or impossible to specify. Experimental application to the problem of risk prediction in pregnancy is described.

Characterisation of the half-field beam penumbra for a variety of blocking set-ups

Measurements of half-field beam penumbra were taken using EBT2 film for a variety of blocking techniques. It was shown that minimizing the SSD reduces the penumbra as the effects of beam divergence are diminished. The addition of a lead block directly on the surface provides optimal results with a 10-90% penumbra of 0.53 ± 0.02 cm. To resolve the uncertainties encountered in film measurements, future Monte Carlo measurements of halffield penumbras are to be conducted.

Using extended random set to find specific patterns

With the overwhelming increase in the amount of data on the web and data bases, many text mining techniques have been proposed for mining useful patterns in text documents. Extracting closed sequential patterns using the Pattern Taxonomy Model (PTM) is one of the pruning methods to remove noisy, inconsistent, and redundant patterns. However, PTM model treats each extracted pattern as whole without considering included terms, which could affect the quality of extracted patterns. This paper propose an innovative and effective method that extends the random set to accurately weigh patterns based on their distribution in the documents and their terms distribution in patterns. Then, the proposed approach will find the specific closed sequential patterns (SCSP) based on the new calculated weight. The experimental results on Reuters Corpus Volume 1 (RCV1) data collection and TREC topics show that the proposed method significantly outperforms other state-of-the-art methods in different popular measures.

“Back to the future” – a retrospective analysis of university business models

Using established strategic management and business model frameworks we map the evolution of universities in the context of their value proposition to students as consumers of their products. We argue that in the main universities over time have transitioned from a value-based business model through to an efficiency-based business model that for numerous reasons, is becoming rapidly unsustainable. We further argue that the future university business models would benefit with a reconfiguration towards a network value based model. This approach requires a revised set of perceived benefits, better aligned to the current and future expectations and an alternate approach to the delivery of those benefits to learner / consumers.

Representing team behaviours from noisy data using player role

Due to their unobtrusive nature, vision-based approaches to tracking sports players have been preferred over wearable sensors as they do not require the players to be instrumented for each match. Unfortunately however, due to the heavy occlusion between players, variation in resolution and pose, in addition to fluctuating illumination conditions, tracking players continuously is still an unsolved vision problem. For tasks like clustering and retrieval, having noisy data (i.e. missing and false player detections) is problematic as it generates discontinuities in the input data stream. One method of circumventing this issue is to use an occupancy map, where the field is discretised into a series of zones and a count of player detections in each zone is obtained. A series of frames can then be concatenated to represent a set-play or example of team behaviour. A problem with this approach though is that the compressibility is low (i.e. the variability in the feature space is incredibly high). In this paper, we propose the use of a bilinear spatiotemporal basis model using a role representation to clean-up the noisy detections which operates in a low-dimensional space. To evaluate our approach, we used a fully instrumented field-hockey pitch with 8 fixed high-definition (HD) cameras and evaluated our approach on approximately 200,000 frames of data from a state-of-the-art real-time player detector and compare it to manually labeled data.

Trailing Touch

Trailing Touch (2014) is a twenty minute classically-based dance work set to music by composers Carter Burwell and Hildur Gudnadottir. It is an abstract work that explores images based on imperfect patterns and seeks to transmit the sensations felt when clusters form and disperse through the space or crisscross to create swirling wave-like reactions in the dancers’ tulle skirts. These simple references are inspired by the lyrical use of arms found in ballet, particular to the ballet aesthetic. Trailing Touch was created in collaboration with QUT’s third-year BFA (Dance Performance) students and performed as part of Dance14 at QUT’s Gardens Points Theatre from the 4th to 8th November 2014 and was performed in Singapore as part of Contact Contemporary Dance Festival on 30th November, 2014. Additionally, the creative process of Trailing Touch (2014) forms the initial project of Phase III of my PhD research, Writing the Dance Score in the Twenty-first Century: An approach for the Independent Choreographer. This PhD research will examine the potential of dance scores as a suite of choreographic strategies to map key aspects of the choreographic process. While a certain degree of ambiguity drives the creative process, the suite of choreographic strategies attempt to capture what is transmitted through the lived experience of dance. “[T]hese documents harbor a force of expression, a visual energy related to the body and the movement” (Louppe 1994, 7) that triggers movement responses, unforeseen intensities and enables personal interpretation. Consequently, Phase III will test and evaluate the relevance of Phase II research within the pressures of mainstream dance rehearsal and performance contexts. In Project One Trailing Touch this was demonstrated in the dance scores produced by the choreographer and interpreted by the dancers within the performance. By drawing from both the theoretical and practical, it is anticipated that this research will suggest a form of languaging movement that is not reliant on images or numbers, but generated in response to the intuitive and complex process underpinning choreographic practice. Rather than constructing a codified dance notation system, it will focus on strategies that reveal movement, its spatial patterns, qualities and intensities of expression and the procedures underlying key choreographic concepts. The outcome of this research project aims to support the independent choreographer in two major areas, by facilitating and enriching the choreographic process for both the performers and choreographer, and by strengthening artistic development and performance outcomes.

Working as one: A road map to disaster resilience for Australia

Natural disasters cause widespread disruption, costing the Australian economy $6.3 billion per year, and those costs are projected to rise incrementally to $23 billion by 2050. With more frequent natural disasters with greater consequences, Australian communities need the ability to prepare and plan for them, absorb and recover from them, and adapt more successfully to their effects. Enhancing Australian resilience will allow us to better anticipate disasters and assist in planning to reduce losses, rather than just waiting for the next king hit and paying for it afterwards. Given the scale of devastation, governments have been quick to pick up the pieces when major natural disasters hit. But this approach (‘The government will give you taxpayers’ money regardless of what you did to help yourself, and we’ll help you rebuild in the same risky area.’) has created a culture of dependence. This is unsustainable and costly. In 2008, ASPI published Taking a punch: building a more resilient Australia. That report emphasised the importance of strong leadership and coordination in disaster resilience policymaking, as well as the value of volunteers and family and individual preparation, in managing the effects of major disasters. This report offers a roadmap for enhancing Australia’s disaster resilience, building on the 2011 National Strategy for Disaster Resilience. It includes a snapshot of relevant issues and current resilience efforts in Australia, outlining key challenges and opportunities. The report sets out 11 recommendations to help guide Australia towards increasing national resilience, from individuals and local communities through to state and federal agencies.

Shortest path and vehicle trajectory aided map-matching for low frequency GPS data

Map-matching algorithms that utilise road segment connectivity along with other data (i.e.position, speed and heading) in the process of map-matching are normally suitable for high frequency (1 Hz or higher) positioning data from GPS. While applying such map-matching algorithms to low frequency data (such as data from a fleet of private cars, buses or light duty vehicles or smartphones), the performance of these algorithms reduces to in the region of 70% in terms of correct link identification, especially in urban and sub-urban road networks. This level of performance may be insufficient for some real-time Intelligent Transport System (ITS) applications and services such as estimating link travel time and speed from low frequency GPS data. Therefore, this paper develops a new weight-based shortest path and vehicle trajectory aided map-matching (stMM) algorithm that enhances the map-matching of low frequency positioning data on a road map. The well-known A* search algorithm is employed to derive the shortest path between two points while taking into account both link connectivity and turn restrictions at junctions. In the developed stMM algorithm, two additional weights related to the shortest path and vehicle trajectory are considered: one shortest path-based weight is related to the distance along the shortest path and the distance along the vehicle trajectory, while the other is associated with the heading difference of the vehicle trajectory. The developed stMM algorithm is tested using a series of real-world datasets of varying frequencies (i.e. 1 s, 5 s, 30 s, 60 s sampling intervals). A high-accuracy integrated navigation system (a high-grade inertial navigation system and a carrier-phase GPS receiver) is used to measure the accuracy of the developed algorithm. The results suggest that the algorithm identifies 98.9% of the links correctly for every 30 s GPS data. Omitting the information from the shortest path and vehicle trajectory, the accuracy of the algorithm reduces to about 73% in terms of correct link identification. The algorithm can process on average 50 positioning fixes per second making it suitable for real-time ITS applications and services.