Exploiting medical hierarchies for concept-based information retrieval

Search technologies are critical to enable clinical sta to rapidly and e ectively access patient information contained in free-text medical records. Medical search is challenging as terms in the query are often general but those in rel- evant documents are very speci c, leading to granularity mismatch. In this paper we propose to tackle granularity mismatch by exploiting subsumption relationships de ned in formal medical domain knowledge resources. In symbolic reasoning, a subsumption (or `is-a') relationship is a parent-child rela- tionship where one concept is a subset of another concept. Subsumed concepts are included in the retrieval function. In addition, we investigate a number of initial methods for combining weights of query concepts and those of subsumed concepts. Subsumption relationships were found to provide strong indication of relevant information; their inclusion in retrieval functions yields performance improvements. This result motivates the development of formal models of rela- tionships between medical concepts for retrieval purposes.

Exploiting SNOMED CT concepts and relationships for clinical information retrieval : Australian e-Health Research Centre and Queensland University of Technology at the TREC 2012 Medical Track

The Australian e-Health Research Centre and Queensland University of Technology recently participated in the TREC 2012 Medical Records Track. This paper reports on our methods, results and experience using an approach that exploits the concept and inter-concept relationships defined in the SNOMED CT medical ontology. Our concept-based approach is intended to overcome specific challenges in searching medical records, namely vocabulary mismatch and granularity mismatch. Queries and documents are transformed from their term-based originals into medical concepts as defined by the SNOMED CT ontology, this is done to tackle vocabulary mismatch. In addition, we make use of the SNOMED CT parent-child `is-a' relationships between concepts to weight documents that contained concept subsumed by the query concepts; this is done to tackle the problem of granularity mismatch. Finally, we experiment with other SNOMED CT relationships besides the is-a relationship to weight concepts related to query concepts. Results show our concept-based approach performed significantly above the median in all four performance metrics. Further improvements are achieved by the incorporation of weighting subsumed concepts, overall leading to improvement above the median of 28% infAP, 10% infNDCG, 12% R-prec and 7% Prec@10. The incorporation of other relations besides is-a demonstrated mixed results, more research is required to determined which SNOMED CT relationships are best employed when weighting related concepts.

Semantic judgement of medical concepts : combining syntagmatic and paradigmatic information with the tensor encoding model

This paper outlines a novel approach for modelling semantic relationships within medical documents. Medical terminologies contain a rich source of semantic information critical to a number of techniques in medical informatics, including medical information retrieval. Recent research suggests that corpus-driven approaches are effective at automatically capturing semantic similarities between medical concepts, thus making them an attractive option for accessing semantic information. Most previous corpus-driven methods only considered syntagmatic associations. In this paper, we adapt a recent approach that explicitly models both syntagmatic and paradigmatic associations. We show that the implicit similarity between certain medical concepts can only be modelled using paradigmatic associations. In addition, the inclusion of both types of associations overcomes the sensitivity to the training corpus experienced by previous approaches, making our method both more effective and more robust. This finding may have implications for researchers in the area of medical information retrieval.

An examination of keystroke dynamics for continuous user authentication

Most current computer systems authorise the user at the start of a session and do not detect whether the current user is still the initial authorised user, a substitute user, or an intruder pretending to be a valid user. Therefore, a system that continuously checks the identity of the user throughout the session is necessary without being intrusive to end-user and/or effectively doing this. Such a system is called a continuous authentication system (CAS). Researchers have applied several approaches for CAS and most of these techniques are based on biometrics. These continuous biometric authentication systems (CBAS) are supplied by user traits and characteristics. One of the main types of biometric is keystroke dynamics which has been widely tried and accepted for providing continuous user authentication. Keystroke dynamics is appealing for many reasons. First, it is less obtrusive, since users will be typing on the computer keyboard anyway. Second, it does not require extra hardware. Finally, keystroke dynamics will be available after the authentication step at the start of the computer session. Currently, there is insufficient research in the CBAS with keystroke dynamics field. To date, most of the existing schemes ignore the continuous authentication scenarios which might affect their practicality in different real world applications. Also, the contemporary CBAS with keystroke dynamics approaches use characters sequences as features that are representative of user typing behavior but their selected features criteria do not guarantee features with strong statistical significance which may cause less accurate statistical user-representation. Furthermore, their selected features do not inherently incorporate user typing behavior. Finally, the existing CBAS that are based on keystroke dynamics are typically dependent on pre-defined user-typing models for continuous authentication. This dependency restricts the systems to authenticate only known users whose typing samples are modelled. This research addresses the previous limitations associated with the existing CBAS schemes by developing a generic model to better identify and understand the characteristics and requirements of each type of CBAS and continuous authentication scenario. Also, the research proposes four statistical-based feature selection techniques that have highest statistical significance and encompasses different user typing behaviors which represent user typing patterns effectively. Finally, the research proposes the user-independent threshold approach that is able to authenticate a user accurately without needing any predefined user typing model a-priori. Also, we enhance the technique to detect the impostor or intruder who may take over during the entire computer session.

Analysis of Cone-Beam CT using prior information

Treatment plans for conformal radiotherapy are based on an initial CT scan. The aim is to deliver the prescribed dose to the tumour, while minimising exposure to nearby organs. Recent advances make it possible to also obtain a Cone-Beam CT (CBCT) scan, once the patient has been positioned for treatment. A statistical model will be developed to compare these CBCT scans with the initial CT scan. Changes in the size, shape and position of the tumour and organs will be detected and quantified. Some progress has already been made in segmentation of prostate CBCT scans [1],[2],[3]. However, none of the existing approaches have taken full advantage of the prior information that is available. The planning CT scan is expertly annotated with contours of the tumour and nearby sensitive objects. This data is specific to the individual patient and can be viewed as a snapshot of spatial information at a point in time. There is an abundance of studies in the radiotherapy literature that describe the amount of variation in the relevant organs between treatments. The findings from these studies can form a basis for estimating the degree of uncertainty. All of this information can be incorporated as an informative prior into a Bayesian statistical model. This model will be developed using scans of CT phantoms, which are objects with known geometry. Thus, the accuracy of the model can be evaluated objectively. This will also enable comparison between alternative models.

Efficient safety information retrieval on construction sites : a preliminary methodology

IT-supported field data management benefits on-site construction management by improving accessibility to the information and promoting efficient communication between project team members. However, most of on-site safety inspections still heavily rely on subjective judgment and manual reporting processes and thus observers’ experiences often determine the quality of risk identification and control. This study aims to develop a methodology to efficiently retrieve safety-related information so that the safety inspectors can easily access to the relevant site safety information for safer decision making. The proposed methodology consists of three stages: (1) development of a comprehensive safety database which contains information of risk factors, accident types, impact of accidents and safety regulations; (2) identification of relationships among different risk factors based on statistical analysis methods; and (3) user-specified information retrieval using data mining techniques for safety management. This paper presents an overall methodology and preliminary results of the first stage research conducted with 101 accident investigation reports.

Efficacy of oral or intrauterine device-delivered progestin in patients with complex endometrial hyperplasia with atypia or early endometrial adenocarcinoma : a meta-analysis and systematic review of the literature

Objectives: To investigate the efficacy of progestin treatment to achieve pathological complete response (pCR) in patients with complex atypical endometrial hyperplasia (CAH) or early endometrial adenocarcinoma (EC). Methods: A systematic search identified 3245 potentially relevant citations. Studies containing less than ten eligible CAH or EC patients in either oral or intrauterine treatment arm were excluded. Only information from patients receiving six or more months of treatment and not receiving other treatments was included. Weighted proportions of patients achieving pCR were calculated using R software. Results: Twelve studies met the selection criteria. Eleven studies reported treatment of patients with oral (219 patients, 117 with CAH, 102 with grade 1 Stage I EC) and one reported treatment of patients with intrauterine progestin (11 patients with grade 1 Stage IEC). Overall, 74% (95% confidence interval [CI] 65-81%) of patients with CAH and 72% (95% CI 62-80%) of patients with grade 1 Stage I EC achieved a pCR to oral progestin. Disease progression while on oral treatment was reported for 6/219 (2.7%), and relapse after initial complete response for 32/159 (20.1%) patients. The weighted mean pCR rate of patients with grade 1 Stage I EC treated with intrauterine progestin from one prospective pilot study and an unpublished retrospective case series from the Queensland Centre of Gynaecologic Oncology (QCGC) was 68% (95% CI 45- 86%). Conclusions: There is a lack of high quality evidence for the efficacy of progestin in CAH or EC. The available evidence however suggests that treatment with oral or intrauterine progestin is similarly effective. The risk of progression during treatment is small but longer follow-up is required. Evidence from prospective controlled clinical trials is warranted to establish how the efficacy of progestin for the treatment of CAH and EC can be improved further.

International students using online information resources to learn : complex experience and learning needs

This paper reports the findings of a qualitative study which investigated 25 international students’ use of online information resources for study purposes at two Australian universities. Using an expanded critical incident approach, the study viewed international students through an information literacy lens, as information-using learners. The findings are presented in two complementary parts: as a word picture that describes their whole experience of using online information resources to learn; and as a tabulated set of critical findings that summarises their associated information literacy learning needs. The word picture shows international students’ resource use as a complex interplay of eight inter-related elements: students; information-learning environment; interactions (with online resources); strengths-challenges; learning-help; affective responses; reflective responses; cultural-linguistic dimensions. In using online resources, the international students experience an array of strengths and challenges, and an apparent information literacy imbalance between their more developed information skills and less developed critical information use. The critical findings about information literacy needs provide a framework for developing an inclusive informed learning approach that responds to international students’ complex information using experiences and needs. While the study is situated in Australia, the findings are of potential interest to educators, information professionals and researchers worldwide who seek to support learning in culturally diverse higher education contexts.

Using information communication technologies to develop dynamic curriculum frameworks for diverse cohorts: a case study from event management

This article investigates the role of information communication technologies (ICTs) in establishing a well-aligned, authentic learning environment for a diverse cohort of non-cognate and cognate students studying event management in a higher education context. Based on a case study which examined the way ICTs assisted in accommodating diverse learning needs, styles and stages in an event management subject offered in the Creative Industries Faculty at Queensland University of Technology in Brisbane, Australia, the article uses an action research approach to generate grounded, empirical data on the effectiveness of the dynamic, individualised curriculum frameworks that the use of ICTs makes possible. The study provides insights into the way non-cognate and cognate students respond to different learning tools. It finds that whilst non-cognate and cognate students do respond to learning tools differently, due to a differing degree of emphasis on technical, task or theoretical competencies, the use of ICTs allows all students to improve their performance by providing multiple points of entry into the content. In this respect, whilst the article focuses on the way ICTs can be used to develop an authentic, well-aligned curriculum model that meets the needs of event management students in a higher education context, with findings relevant for event educators in Business, Hospitality, Tourism and Creative Industries, the strategies outlined may also be useful for educators in other fields who are faced with similar challenges when designing and developing curriculum for diverse cohorts.

Cross layer optimization of VANET routing with multi-objective decision making

Vehicular Ad-hoc Networks (VANET) have different characteristics compared to other mobile ad-hoc networks. The dynamic nature of the vehicles which act as routers and clients are connected with unreliable radio links and Routing becomes a complex problem. First we propose CO-GPSR (Cooperative GPSR), an extension of the traditional GPSR (Greedy Perimeter Stateless Routing) which uses relay nodes which exploit radio path diversity in a vehicular network to increase routing performance. Next we formulate a Multi-objective decision making problem to select optimum packet relaying nodes to increase the routing performance further. We use cross layer information for the optimization process. We evaluate the routing performance more comprehensively using realistic vehicular traces and a Nakagami fading propagation model optimized for highway scenarios in VANETs. Our results show that when Multi-objective decision making is used for cross layer optimization of routing a 70% performance increment can be obtained for low vehicle densities on average, which is a two fold increase compared to the single criteria maximization approach.

Exploring the information practices of people with end stage kidney disease

This small exploratory study sought to understand how people with End Stage Kidney Disease (ESKD) experience the information environment and what information practices they employ in order to inform the decisions they make in relation to treatment and care. Using a constructivist methodology, in-depth interviews were conducted with five people who were receiving haemodialysis in two small satellite dialysis units located in regional and rural communities in New South Wales, Australia. Thematic analysis revealed two types of patients. The first type appears to adopt a received view of information, who do not question their condition; and passively accept information. In the other type, patients were found to be engaged; they actively identified their information needs and quickly learned what that they needed to ask and who to ask. Knowing the information practices of people with ESKD is useful for nephrology nurses when providing patient education.

What information-related activities do people with ESKD use?

Background Information practice is an emerging area of research that seeks to reveal how people learn to connect with the complex multimodal information landscapes that informs their ability to make decisions. Previous research has identified that people with end stage kidney disease (ESKD) tend to adopt a ‘received’ or ‘engaged’ view of information but little is known about the activities of information practice. Objectives This research project sought to identify the: i) information-related activities; and ii) how information is used. Methods Using a constructivist qualitative methodology, ten people with ESKD living in a large metropolitan city were purposively selected and interviewed. Data was subject to thematic analysis by researchers from nursing and information science. Saturation of themes was achieved. Results Participants were between 38 and 72 years, had been receiving kidney replacement therapy from 2 weeks to 31 years. Eight participants reported having access to the internet but none participated in chat rooms. The activities were conceptualized into themes as listening, seeking, searching, sharing and observing. These activities enabled people to create, reflect on and evaluate the information needed to inform their decision-making Conclusion/Application to Clinical Practice The information practice research approach will enable a better understanding of the underlying relationship between information, knowledge and experience to be better understood. For renal nurses who are involved in patient education being able to recognise the way people use information will assist in individualizing educational sessions and tailoring teaching strategies to make it more meaningful.

Emergency Health Services (EHS) : demand and service delivery models. [Monograph 2 : Queensland EHS users’ profile]

Emergency health is a critical component of Australia’s health system and one which is increasingly congested from growing demand and blocked access to inpatient beds. The Emergency Health Services Queensland (EHSQ) study aims to identify the factors driving increased demand for emergency health and to evaluate strategies which may safely reduce the future demand growth. This monograph addresses the characteristics of users of emergency health services with an aim to identify those that appear to contribute to demand growth. This study utilises data on patients treated by Emergency Departments (ED) and Queensland Ambulance Service (QAS) across Queensland. ED data was derived from the Emergency Department Information System (EDIS) for the period 2001-02 through to 2010-11. Ambulance data was extracted from the QAS’ Ambulance Information Management System (AIMS) and electronic Ambulance Report Form (eARF) for the period 2001-02 through to 2009-10. Due to discrepancies and comparability issues for ED data, this monograph compares data from the 2003-04 time period with 2010-11 data for 21 of the reporting EDs. Also a snapshot of users for the 2010-11 financial year for 31 reporting EDs is used to describe the characteristics of users and to compare those characteristics with population demographics. For QAS data, the 2002-03 and 2009-10 time periods were selected for detailed analyses to identify trends. • Demand for emergency health care services is increasing, representing both increased population and increased relative utilisation. Per capita demand for ED attention has increased by 2% per annum over the last decade and for ambulance attention by 3.7% per annum. • The growth in ED demand is prominent in more urgent triage categories with actual decline in less urgent patients. An estimated 55% of patients attend hospital EDs outside of normal working hours. There is no evidence that patients presenting out of hours are significantly different to those presenting within working hours; they have similar triage assessments and outcomes. • Patients suffering from injuries and poisoning comprise 28% of the ED workload (an increase of 65% in the study period), whilst declines of 32% in cardiovascular and circulatory conditions, and musculoskeletal problems have been observed. • 25.6% of patients attending EDs are admitted to hospital. 19% of admitted patients and 7% of patients who die in the ED are triage category 4 or 5 on arrival. • The average age of ED patients is 35.6 years. Demand has grown in all age groups and amongst both men and women. Men have higher utilisation rates for ED in all age groups. The only group where the growth rate in women has exceeded men is in the 20-29 age group; this growth is particularly in the injury and poisoning categories. • Considerable attention has been paid publicly to ED performance criteria. It is worth noting that 50% of all patients were treated within 33 minutes of arrival. • Patients from lower socioeconomic areas appear to have higher utilisation rates and the utilisation rate for indigenous people appears to exceed those of European and other backgrounds. The utilisation rates for immigrant people is generally less than that of Australian born however it has not been possible to eliminate the confounding impact of different age and socioeconomic profiles. • Demand for ambulance service is also increasing at a rate that exceeds population growth. Utilisation rates have increased by an average of 5% per annum in Queensland compared to 3.6% nationally, and the utilisation rate in Queensland is 27% higher than the national average. • The growth in ambulance utilisation has also been amongst the more urgent categories of dispatch and utilisation rates are higher in rural and regional areas than in the metropolitan area. The demand for ambulance increases with age but the growth in demand for ambulance service has been more prominent in younger age groups. These findings contribute significantly to an understanding of the growth in demand for emergency health. It shows that the growth is amongst patients in genuine need of emergency healthcare and public rhetoric that the congestion of emergency health services is due to inappropriate attendees is unable to be substantiated. The consistency of the growth in demand over the last decade reflects not only the changing demographics of the Australian population but also the changes in health status, standards of acute health care and other social factors. The growth is also amongst patients with acute injury and poisoning which is inconsistent with rates of chronic disease as a fundamental driver. We have also interviewed patients in regard to their decision making choices for acute health care and the factors that influence these decisions and this will be the subject of a third Monograph and publications.

Learning personalized tag ontology from user tagging information

The cross-sections of the Social Web and the Semantic Web has put folksonomy in the spot light for its potential in overcoming knowledge acquisition bottleneck and providing insight for "wisdom of the crowds". Folksonomy which comes as the results of collaborative tagging activities has provided insight into user's understanding about Web resources which might be useful for searching and organizing purposes. However, collaborative tagging vocabulary poses some challenges since tags are freely chosen by users and may exhibit synonymy and polysemy problem. In order to overcome these challenges and boost the potential of folksonomy as emergence semantics we propose to consolidate the diverse vocabulary into a consolidated entities and concepts. We propose to extract a tag ontology by ontology learning process to represent the semantics of a tagging community. This paper presents a novel approach to learn the ontology based on the widely used lexical database WordNet. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users’ tagging behavior together. We provide empirical evaluations by using the semantic information contained in the ontology in a tag recommendation experiment. The results show that by using the semantic relationships on the ontology the accuracy of the tag recommender has been improved.