The aim of this paper is to investigate semantic web based methods to enrich and transform a medical discussion forum in order to perform semantics-driven social network analysis. We use the centrality measures as well as semantic similarity metrics to identify the most influential practitioners within a discussion forum. The centrality results of our approach are in line with centrality measures produced by traditional SNA methods, thus validating the applicability of semantic web based methods for SNA, particularly for analyzing social networks for specialized discussion forums.
Christopher Munro, Philip Couch, Jon Johnson, John Ainsworth, Iain Buchan
491 - 495
Discovery of useful relationships between scholarly assets on the web is challenging, both in terms generating the right metadata around the assets, and in connecting all relevant digital entities in chain of provenance accessible to the whole community. This paper reports the development of a framework and tools enabling scholarly asset relationships to be expressed in a standard and open way, illustrated with use-cases of discovering new knowledge across cohort studies. The framework uses Research Objects for aggregation, distributed databases for storage, and distributed ledgers for provenance. Our proposal avoids management by a single central platform or organization, instead leveraging the use of existing resources and platforms across natural partnerships. Our proposed infrastructure will support a wide range of users from system administrators to researchers.
Allen J. Flynn, Namita Bahulekar, Peter Boisvert, Carl Lagoze, George Meng, James Rampton, Charles P. Friedman
496 - 500
Throughout the world, biomedical knowledge is routinely generated and shared through primary and secondary scientific publications. However, there is too much latency between publication of knowledge and its routine use in practice. To address this latency, what is actionable in scientific publications can be encoded to make it computable. We have created a purpose-built digital library platform to hold, manage, and share actionable, computable knowledge for health called the Knowledge Grid Library. Here we present it with its system architecture.
With the unprecedented increase of healthcare data, technologies need to be both, highly efficient for the meaningful utilization of accessible data and flexible to adapt to future challenges and individual preferences. The OntoHealth system makes use of semantic technologies to enable flexible and individual interaction with Electronic Health Records (EHR) for physicians. This is achieved by the execution of formally modelled clinical workflows and the composition of Semantic Web Services (SWS). Several seamless components provide a service-oriented structure defined by individual designed EHR-workflows. This work gives an overview of the planned architecture and its main components. The architecture constitutes the basis for the prototype implementation of all components. With its highly dynamic structure based on SWS, the architecture will be able to cope with both, the individual users' needs as well as the quick evolving healthcare domain.
Yasmin Van Kasteren, Patricia A.H. Williams, Anthony Maeder
506 - 510
Medical informatics is a young and rapidly evolving field, influenced by and impacting on many different knowledge domains. Recent contributions on scoping the associated body of knowledge are confounded both by variations in popular use of terminology for established areas, and by the advent of new areas without yet established terminology. Determining the scope of a topic through online bibliographic search filters is a well-established approach in scientific research and has been developed as a human-directed task. Establishing the best approach and automating the process has proved a difficult problem. This paper explores the use of text analysis of bibliographic information using available search engines and NVIVO text analysis tools to test the potential for dynamic word based filters based on data mining. Results show that word searches of abstracts are more effective than topic searches for identifying health informatics papers, however more work is required to refine search terms to improve generalisability. Using data mining to track changes in word use in medical informatics journals, may make it possible to establish a more dynamic search filter to match the evolving nature of the field of health informatics.
Children are dependent on a reliable healthcare system, especially for the delivery of care which crosses the primary/secondary care boundary. A methodology based on UML has been developed to capture and single out meaningful parts of the child healthcare pathways in order to facilitate comparison among 30 EU countries within the MOCHA project. A first application of this methodology has been reported considering asthma management as an example.
Mercedes Arguello Casteleiro, Diego Maseda Fernandez, George Demetriou, Warren Read, Maria Jesus Fernandez Prieto, Julio Des Diz, Goran Nenadic, John Keane, Robert Stevens
516 - 520
We investigate the application of distributional semantics models for facilitating unsupervised extraction of biomedical terms from unannotated corpora. Term extraction is used as the first step of an ontology learning process that aims to (semi-)automatic annotation of biomedical concepts and relations from more than 300K PubMed titles and abstracts. We experimented with both traditional distributional semantics methods such as Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation (LDA) as well as the neural language models CBOW and Skip-gram from Deep Learning. The evaluation conducted concentrates on sepsis, a major life-threatening condition, and shows that Deep Learning models outperform LSA and LDA with much higher precision.
Technical medical terms are complicated to be correctly understood by non-experts. Vocabulary, associating technical terms with layman expressions, can help in increasing the readability of technical texts and their understanding. The purpose of our work is to build this kind of vocabulary. We propose to exploit the notion of reformulation following two methods: extraction of abbreviations and of reformulations with specific markers. The segments associated thanks to these methods are aligned with medical terminologies. Our results allow to cover over 9,000 medical terms and show precision of extractions between 0.24 and 0.98. The results and analyzed and compared with the existing work.
Non-optimal prescriptions of antibiotics have a negative impact on patients and population. Clinical practice guidelines are not always followed by doctors because the rationale of the recommendations is not always clear and can be difficult to understand. In this paper, we propose a new approach consisting in presenting the properties of antibiotics for allowing doctors to compare them and choose the most appropriate one. For that, we used and extended rainbow boxes, a new technique for overlapping set visualization. We tested our approach on 11 clinical situations related to urinary infections, and assessed the simplicity, the interest and utility with 11 doctors. 10 of them found that this approach was interesting and useful in clinical practice. Further studies are needed to confirm this preliminary work.
Over the past twenty-five years the time from diagnosis of breast cancer to the initiation of therapy has steadily grown. In this note we present a mechanism to give a ballpark estimate of the risk associated with delaying therapy given a specific set of presenting patient data.
Carlos Sáez, David Moner, Ricardo García-De-León-Chocano, Verónica Muñoz-Soler, Ricardo García-De-León-González, José Alberto Maldonado, Diego Boscá, Salvador Tortajada, Montserrat Robles, Juan M. García-Gómez, Manuel Alcaraz, Pablo Serrano, José L. Bernal, Jesús Rodríguez, Gerardo Bustos, Miguel Esparza
539 - 543
We present the results of a pilot project of the Spanish Ministry of Health, Social Services and Equality, envisaged to the development of a national integrated data repository of maternal-child care information. Based on health information standards and data quality assessment procedures, the developed repository is aimed to a reliable data reuse for (1) population research and (2) the monitoring of healthcare best practices. Data standardization was provided by means of two main ISO 13606 archetypes (composed of 43 sub-archetypes), the first dedicated to the delivery and birth information and the second about the infant feeding information from delivery up to two years. Data quality was assessed by means of a dedicated procedure on seven dimensions including completeness, consistency, uniqueness, multi-source variability, temporal variability, correctness and predictive value. A set of 127 best practice indicators was defined according to international recommendations and mapped to the archetypes, allowing their calculus using XQuery programs. As a result, a standardized and data quality assessed integrated data respository was generated, including 7857 records from two Spanish hospitals: Hospital Virgen del Castillo, Yecla, and Hospital 12 de Octubre, Madrid. This pilot project establishes the basis for a reliable maternal-child care data reuse and standardized monitoring of best practices based on the developed information and data quality standards.
The Medical Research Council (MRC) framework for complex interventions provides useful guidance to assist with the development and evaluation of health technology interventions such as decision support. In this paper we briefly summarise a project that focused on designing a decision support intervention to assist with the recognition, assessment and management of pain in patients with dementia in an acute hospital setting. We reflect on our experience of using the MRC framework to guide our study design, and highlight the importance of considering decision support interventions as complex interventions.
Carsten Oliver Schmidt, Christine Krabbe, Janka Schössow, Martin Albers, Dörte Radke, Jörg Henke
549 - 553
Valid scientific inferences from epidemiological and clinical studies require high data quality. Data generating departments therefore aim to detect data irregularities as early as possible in order to guide quality management processes. In addition, after the completion of data collections the obtained data quality must be evaluated. This can be challenging in complex studies due to a wide scope of examinations, numerous study variables, multiple examiners, devices, and examination centers. This paper describes a Java EE web application used to monitor and evaluate data quality in institutions with complex and multiple studies, named Square2. It uses the Java libraries Apache MyFaces 2, extended by BootsFaces for layout and style. RServe and REngine manage calls to R server processes. All study data and metadata are stored in PostgreSQL. R is the statistics backend and LaTeX is used for the generation of print ready PDF reports. A GUI manages the entire workflow. Square2 covers all steps in the data monitoring workflow, including the setup of studies and their structure, the handling of metadata for data monitoring purposes, selection of variables, upload of data, statistical analyses, and the generation as well as inspection of quality reports. To take into account data protection issues, Square2 comprises an extensive user rights and roles concept.
Ibrahim Habli, Sean White, Stuart Harrison, Manpreet Pujara
554 - 558
Safety analysis is centred on identifying a set of hazards that form the basis of risk assessment. In healthcare, hazards are potential sources of harm to patients and as such the risk of these has to be assessed and managed. With the increased reliance on Health IT systems in health and social care settings, some of these hazards are associated with the development and use of these systems. In this paper we examine current practices in hazard identification, focusing on how clinicians and engineers approach this task within the Health IT safety assurance process. We highlight certain technical and organisational challenges and discuss approaches to improving current practices and promoting learning initiatives.
This paper discusses reactive improvement of clinical software using methods for incident analysis. We used the “Five Whys” method because we had only descriptive data and depended on a domain expert for the analysis. The analysis showed that there are two major root causes for EHR software failure, and that they are related to human and organizational errors. A main identified improvement is allocating more resources to system maintenance and user training.
The transfer of information and responsibility for care of a patient from one healthcare provider to another is referred to as a handover. While some handovers are effective and achieve high quality communication, others represent a barrier to continuity of care. To increase the patient safety, Norway decided to replace handovers with an electronic e-message system (EMS). This paper refers to a quantitative study of this implementation and examines the opinions of first-line leaders and nurses (N = 108) on how organisational factors were taken into account and how the implementation might be improved. The findings indicate that such factors generally did not receive very much attention in the implementation of the EMS, and less for the nurses than for the first-line leaders. Particularly, the factor most prominently identified by both groups as warranted improvement, was the training.
Potentially inappropriate prescribing is a common problem, especially in elderly care. To tackle this problem, Irish medical experts have developed a list of criteria when medication should be added or omitted based upon the patient's physical condition and medication use, known as the STOPP and START criteria. The STOPP and START criteria have been formulated to identify the prescribing of potentially inappropriate medicines (PIMs) and potential prescribing omissions (PPOs). One of the most common problems of inappropriate prescribing is gastro-intestinal track bleedings. For this purpose, nine of the 87 STOPP and START criteria are designed to prevent this. However, the prevalence of gastro-intestinal track bleedings has not been established when these nine STOPP and START criteria are violated. The database contained 182,000 patients belonging to 49 general practitioners in the region of Amsterdam, The Netherlands. We estimated both the incidence of PIMs and PPOs and whether harm, in this case a gastro-intestinal track bleeding, occurred. We found that although violation of the nine STOPP or START criteria were possibly associated with harm (OR = 1.30), this association was not statistically significant (p = 0.323). Searching for evidence for harm informs decision support design aimed at improving quality of medication prescription as it prioritizes the many suggested criteria based on their relevance.
Jamison D. Fargo, Ann Elizabeth Montgomery, Thomas Byrne, Emily Brignone, Meagan Cusack, Adi V. Gundlapalli
574 - 578
Effectiveness of screening for homelessness in a large healthcare system was evaluated in terms of successfully referring and connecting patients with appropriate prevention or intervention services. Screening and healthcare services data from nearly 6 million U.S. military veterans were analyzed. Veterans either screened positive for current or risk of housing instability, or negative for both. Current living situation was used to validate results of screening. Administrative evidence for homelessness-related services was significantly higher among positive-screen veterans who accepted a referral for services compared to those who declined. Screening for current or risk of homelessness led to earlier identification, which led to earlier and more extensive service engagement.
This paper outlines a systematic literature review undertaken to establish current evidence regarding the impact of Business Intelligence (BI) on health system decision making and organizational performance. The review also examined BI implementation factors contributing to these constructs. Following the systematic review, inductive content analysis was used to categorize themes within the eight articles identified. This study demonstrated there is little evidence based literature focused on BI impact on organizational decision making and performance within health care. There was evidence found that BI does improve decision making. Implementation success was found to be dependent on several factors, many of which relate to broader organizational culture and readiness.
Marie-José Roos-Blom, Wouter T. Gude, Evert de Jonge, Jan Jaap Spijkstra, Sabine N. van der Veer, Dave A. Dongelmans, Nicolette F. de Keizer
584 - 588
Audit and feedback (A&F) is a common strategy to improve quality of care. Meta-analyses have indicated that A&F may be more effective in realizing desired change when baseline performance is low, it is delivered by a supervisor or colleague, it is provided frequently and in a timely manner, it is delivered in both verbal and written formats, and it includes specific targets and an action plan. However, there is little information to guide operationalization of these factors. Researchers have consequently called for A&F interventions featuring well-described and carefully justified components, with their theoretical rationale made explicit. This paper describes the rationale and development of a quality dashboard including an improvement toolbox for four previous developed pain indicators, guided by Control Theory.
Samina Abidi, Michael Vallis, Helena Piccinini-Vallis, Syed Ali Imran, Syed Sibte Raza Abidi
589 - 593
We present Diabetes Web-Centric Information and Support Environment (D-WISE) that features: (a) Decision support tool to assist family physicians to administer Behavior Modification (BM) strategies to patients; and (b) Patient BM application that offers BM strategies and motivational interventions to engage patients. We take a knowledge management approach, using semantic web technologies, to model the social cognition theory constructs, Canadian diabetes guidelines and BM protocols used locally, in terms of a BM ontology that drives the BM decision support to physicians and BM strategy adherence monitoring and messaging to patients. We present the qualitative analysis of D-WISE usability by both physicians and patients